{\rtf1\adeflang1025\ansi\ansicpg936\uc2\adeff0\deff0\stshfdbch13\stshfloch0\stshfhich0\stshfbi0\deflang1033\deflangfe2052{\fonttbl{\f0\froman\fcharset0\fprq2{\*\panose 02020603050405020304}Times New Roman;}{\f13\fnil\fcharset134\fprq2{\*\panose 02010600030101010101}\'cb\'ce\'cc\'e5{\*\falt SimSun};} {\f18\fmodern\fcharset136\fprq1{\*\panose 02020309000000000000}MingLiU{\*\falt \'b2\'d3\'a9\'fa\'c5\'e9};}{\f36\fnil\fcharset134\fprq2{\*\panose 02010600030101010101}@\'cb\'ce\'cc\'e5;} {\f82\fmodern\fcharset136\fprq1{\*\panose 02020309000000000000}@MingLiU;}{\f179\froman\fcharset238\fprq2 Times New Roman CE;}{\f180\froman\fcharset204\fprq2 Times New Roman Cyr;}{\f182\froman\fcharset161\fprq2 Times New Roman Greek;} {\f183\froman\fcharset162\fprq2 Times New Roman Tur;}{\f184\fbidi \froman\fcharset177\fprq2 Times New Roman (Hebrew);}{\f185\fbidi \froman\fcharset178\fprq2 Times New Roman (Arabic);}{\f186\froman\fcharset186\fprq2 Times New Roman Baltic;} {\f187\froman\fcharset163\fprq2 Times New Roman (Vietnamese);}{\f311\fnil\fcharset0\fprq2 SimSun Western{\*\falt SimSun};}{\f361\fmodern\fcharset0\fprq1 MingLiU Western{\*\falt \'b2\'d3\'a9\'fa\'c5\'e9};} {\f541\fnil\fcharset0\fprq2 @\'cb\'ce\'cc\'e5 Western;}{\f1001\fmodern\fcharset0\fprq1 @MingLiU Western;}}{\colortbl;\red0\green0\blue0;\red0\green0\blue255;\red0\green255\blue255;\red0\green255\blue0;\red255\green0\blue255;\red255\green0\blue0; \red255\green255\blue0;\red255\green255\blue255;\red0\green0\blue128;\red0\green128\blue128;\red0\green128\blue0;\red128\green0\blue128;\red128\green0\blue0;\red128\green128\blue0;\red128\green128\blue128;\red192\green192\blue192;\red255\green153\blue0;} {\stylesheet{\qj \li0\ri0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\f0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 \snext0 Normal;} {\*\cs10 \additive \ssemihidden Default Paragraph Font;}{\* \ts11\tsrowd\trftsWidthB3\trpaddl108\trpaddr108\trpaddfl3\trpaddft3\trpaddfb3\trpaddfr3\trcbpat1\trcfpat1\tblind0\tblindtype3\tscellwidthfts0\tsvertalt\tsbrdrt\tsbrdrl\tsbrdrb\tsbrdrr\tsbrdrdgl\tsbrdrdgr\tsbrdrh\tsbrdrv \ql \li0\ri0\widctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs20 \ltrch\fcs0 \fs20\lang1024\langfe1024\loch\f0\hich\af0\dbch\af13\cgrid\langnp1024\langfenp1024 \snext11 \ssemihidden Normal Table;}{ \s15\qc \li0\ri0\nowidctlpar\brdrb\brdrs\brdrw15\brsp20 \tqc\tx4153\tqr\tx8306\wrapdefault\aspalpha\aspnum\faauto\nosnaplinegrid\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs18\alang1025 \ltrch\fcs0 \fs18\lang1033\langfe2052\kerning2\loch\f0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 \sbasedon0 \snext15 \styrsid15098623 header;}{\s16\ql \li0\ri0\nowidctlpar \tqc\tx4153\tqr\tx8306\wrapdefault\aspalpha\aspnum\faauto\nosnaplinegrid\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs18\alang1025 \ltrch\fcs0 \fs18\lang1033\langfe2052\kerning2\loch\f0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 \sbasedon0 \snext16 \styrsid15098623 footer;}{\*\cs17 \additive \rtlch\fcs1 \af0 \ltrch\fcs0 \ul\cf2 \sbasedon10 \styrsid15098623 Hyperlink;}}{\*\latentstyles\lsdstimax156\lsdlockeddef0}{\*\pgptbl {\pgp\ipgp0\itap0\li0\ri0\sb0\sa0}}{\*\rsidtbl \rsid655764 \rsid811358\rsid1069012\rsid1074055\rsid1318786\rsid1462542\rsid2231290\rsid2847010\rsid2909350\rsid3146529\rsid3152368\rsid3153644\rsid3409549\rsid3869496\rsid4001084\rsid4197693\rsid4419642\rsid4477090\rsid5848812\rsid6441775\rsid6492656\rsid6493368 \rsid7939098\rsid8674677\rsid8869018\rsid8944438\rsid9449096\rsid10106844\rsid10236727\rsid10246904\rsid10505668\rsid10707375\rsid10897542\rsid11282186\rsid11941893\rsid12669821\rsid13312263\rsid13773176\rsid14707564\rsid14892288\rsid15098623\rsid15222873 \rsid15611782\rsid15665703\rsid15817367\rsid15866361\rsid16080730\rsid16195404\rsid16517661\rsid16546564\rsid16607898}{\*\generator Microsoft Word 11.0.0000;}{\info{\title Pajek data sets, V. Batagelj and A. Mrvar.}{\subject Pajek data sets, V. Batagelj and A. Mrvar.}{\author Document Search} {\keywords Document Search}{\doccomm http://www.downhi.com/txt/PMhieofiFTRK.html}{\operator www.downhi.com}{\creatim\yr2010\mo9\dy28\hr22\min9}{\revtim\yr2014\mo4\dy4\hr1\min23}{\version26}{\edmins1077}{\nofpages1}{\nofwords27}{\nofchars168}{\*\manager http://www.downhi.com/} {\*\company http://www.downhi.com/}{\*\category Document Search}{\nofcharsws186}{\vern24617}{\*\password 00000000}}{\*\xmlnstbl {\xmlns1 http://schemas.microsoft.com/office/word/2003/wordml}{\xmlns2 urn:schemas-microsoft-com:office:smarttags}} \paperw11906\paperh16838\margl1134\margr1134\margt1134\margb1134\gutter0\ltrsect \deftab420\ftnbj\aenddoc\donotembedsysfont1\donotembedlingdata0\grfdocevents0\validatexml1\showplaceholdtext0\ignoremixedcontent0\saveinvalidxml0\showxmlerrors1\formshade\horzdoc\dgmargin\dghspace180\dgvspace156\dghorigin1134\dgvorigin1134\dghshow0 \dgvshow2\jcompress\lnongrid\viewkind1\viewscale85\splytwnine\ftnlytwnine\htmautsp\useltbaln\alntblind\lytcalctblwd\lyttblrtgr\lnbrkrule\nobrkwrptbl\snaptogridincell\allowfieldendsel\wrppunct\asianbrkrule\rsidroot3869496\newtblstyruls\nogrowautofit {\*\fchars !),.:\'3b?]\'7d\'a1\'a7\'a1\'a4\'a1\'a6\'a1\'a5\'a8\'44\'a1\'ac\'a1\'af\'a1\'b1\'a1\'ad\'a1\'c3\'a1\'a2\'a1\'a3\'a1\'a8\'a1\'a9\'a1\'b5\'a1\'b7\'a1\'b9\'a1\'bb\'a1\'bf\'a1\'b3\'a1\'bd\'a3\'a1\'a3\'a2\'a3\'a7\'a3\'a9\'a3\'ac\'a3\'ae\'a3\'ba\'a3\'bb\'a3\'bf\'a3\'dd\'a3\'e0\'a3\'fc\'a3\'fd\'a1\'ab\'a1\'e9 }{\*\lchars ([\'7b\'a1\'a4\'a1\'ae\'a1\'b0\'a1\'b4\'a1\'b6\'a1\'b8\'a1\'ba\'a1\'be\'a1\'b2\'a1\'bc\'a3\'a8\'a3\'ae\'a3\'db\'a3\'fb\'a1\'ea\'a3\'a4}\fet0{\*\wgrffmtfilter 013f}\ilfomacatclnup0{\*\template C:\\Documents and Settings\\Administrator\\\'d7\'c0\'c3\'e6\\doc.dot}{\*\ftnsep \ltrpar \pard\plain \ltrpar\qj \li0\ri0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid14707564 \chftnsep \par }}{\*\ftnsepc \ltrpar \pard\plain \ltrpar\qj \li0\ri0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid14707564 \chftnsepc \par }}{\*\aftnsep \ltrpar \pard\plain \ltrpar\qj \li0\ri0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid14707564 \chftnsep \par }}{\*\aftnsepc \ltrpar \pard\plain \ltrpar\qj \li0\ri0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid14707564 \chftnsepc \par }}\ltrpar \sectd \ltrsect\linex0\headery851\footery992\colsx425\endnhere\sectlinegrid312\sectspecifyl\sectrsid6493368\sftnbj {\headerr \ltrpar \pard\plain \ltrpar\s15\qc \li0\ri0\nowidctlpar\brdrb\brdrs\brdrw15\brsp20 \tqc\tx4153\tqr\tx8306\wrapdefault\aspalpha\aspnum\faauto\nosnaplinegrid\adjustright\rin0\lin0\itap0 \rtlch\fcs1 \af0\afs18\alang1025 \ltrch\fcs0 \fs18\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0\afs30 \ltrch\fcs0 \b\f13\fs30\cf6\insrsid1074055\charrsid1074055 \hich\af13\dbch\af13\loch\f13 Free Document Search and Download}{\rtlch\fcs1 \af0\afs30 \ltrch\fcs0 \b\fs30\cf6\loch\af13\insrsid6493368\charrsid1074055 \par }{\field{\*\fldinst {\rtlch\fcs1 \af0\afs32 \ltrch\fcs0 \f13\fs32\cf6\insrsid14892288 \hich\af13\dbch\af13\loch\f13 HYPERLINK "http://www.downhi.com/" }{\rtlch\fcs1 \af0\afs32 \ltrch\fcs0 \fs32\cf6\loch\af13\insrsid10707375\charrsid14892288 {\*\datafield 00d0c9ea79f9bace118c8200aa004ba90b0200000003000000e0c9ea79f9bace118c8200aa004ba90b4e00000068007400740070003a002f002f00770065006e00640061006e0067002e0064006f00630073006f0075002e0063006f006d002f000000795881f43b1d7f48af2c825dc485276300000000a5ab0000000000}} }{\fldrslt {\rtlch\fcs1 \af0\afs32 \ltrch\fcs0 \cs17\f13\fs32\ul\cf2\insrsid3146529\charrsid14892288 \hich\af13\dbch\af13\loch\f13 http://www.downhi.com/}}}\sectd \linex0\endnhere\sectdefaultcl\sftnbj {\rtlch\fcs1 \af0\afs32 \ltrch\fcs0 \fs32\cf6\loch\af13\insrsid6493368\charrsid15098623 \par }}{\*\pnseclvl1\pnucrm\pnstart1\pnindent720\pnhang {\pntxta \dbch .}}{\*\pnseclvl2\pnucltr\pnstart1\pnindent720\pnhang {\pntxta \dbch .}}{\*\pnseclvl3\pndec\pnstart1\pnindent720\pnhang {\pntxta \dbch .}}{\*\pnseclvl4\pnlcltr\pnstart1\pnindent720\pnhang {\pntxta \dbch )}}{\*\pnseclvl5\pndec\pnstart1\pnindent720\pnhang {\pntxtb \dbch (}{\pntxta \dbch )}}{\*\pnseclvl6\pnlcltr\pnstart1\pnindent720\pnhang {\pntxtb \dbch (}{\pntxta \dbch )}}{\*\pnseclvl7\pnlcrm\pnstart1\pnindent720\pnhang {\pntxtb \dbch (} {\pntxta \dbch )}}{\*\pnseclvl8\pnlcltr\pnstart1\pnindent720\pnhang {\pntxtb \dbch (}{\pntxta \dbch )}}{\*\pnseclvl9\pnlcrm\pnstart1\pnindent720\pnhang {\pntxtb \dbch (}{\pntxta \dbch )}}\pard\plain \ltrpar\qj \li0\ri0\sl180\slmult0 \nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid6493368 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid6493368\charrsid1074055 \loch\af0\hich\af0\dbch\f13 \'a1\'a1\'a1\'a1}{\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid6493368\charrsid1074055 \par }\pard \ltrpar\qc \li0\ri0\sl180\slmult0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid1318786 {\rtlch\fcs1 \af0\afs36 \ltrch\fcs0 \b\fs36\insrsid1318786\charrsid1074055 \hich\af0\dbch\af13\loch\f0 Pajek data sets, V. Batagelj and A. Mrvar. \par }\pard \ltrpar\qj \li0\ri0\sl180\slmult0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid1074055 {\rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid6493368\charrsid1074055 \loch\af0\hich\af0\dbch\f13 \'a3\'ba}{\rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid6493368\charrsid1074055 \par }\pard \ltrpar\qc \li0\ri0\sl180\slmult0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid3869496 {\field{\*\fldinst {\rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid1074055 \hich\af0\dbch\af13\loch\f0 \hich\af0\dbch\af13\loch\f0 HYPERLINK "http://www.downhi.com/"\hich\af0\dbch\af13\loch\f0 }{\rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid13719882\charrsid1074055 {\*\datafield 00d0c9ea79f9bace118c8200aa004ba90b0200000003000000e0c9ea79f9bace118c8200aa004ba90b4600000068007400740070003a002f002f007700770077002e0064006f0077006e00680069002e0063006f006d002f000000795881f43b1d7f48af2c825dc485276300000000a5ab0000}}}{\fldrslt { \rtlch\fcs1 \af0 \ltrch\fcs0 \cs17\fs24\ul\cf2\insrsid3869496\charrsid1074055 \hich\af0\dbch\af13\loch\f0 http://www.downhi.com/txt/PMhieofiFTRK.html}{\rtlch\fcs1 \af0 \ltrch\fcs0 \cs17\fs24\ul\cf2\insrsid6493368\charrsid1074055 \par }\pard \ltrpar\qj \li0\ri0\sl180\slmult0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid6493368 }}\pard\plain \ltrpar\qj \li0\ri0\sl180\slmult0 \nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid6493368 \rtlch\fcs1 \af0\afs24\alang1025 \ltrch\fcs0 \fs21\lang1033\langfe2052\kerning2\loch\af0\hich\af0\dbch\af13\cgrid\langnp1033\langfenp2052 \sectd \linex0\headery851\footery992\colsx425\endnhere\sectlinegrid312\sectspecifyl\sectrsid6493368\sftnbj {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid6493368\charrsid1074055 \par }{\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid6493368\charrsid1074055 \par }\pard \ltrpar\qj \li0\ri0\sl360\slmult1\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid15866361 {\rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid3869496\charrsid1074055 \hich\af0\dbch\af13\loch\f0 Pajek data sets, V. Batagelj and A. Mrvar.\par \par Pajek data sets, from http://vlado.fmf.uni-lj.si/pub/networks/data/,\par Vladimir Batagelj and Andrej Mrvar (2006): Pajek datasets. If the source of\par the data set is not specified otherwise, these data sets are licensed under a\par Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License.\par \par Converted to sparse adjacency matrix format by Tim Davis, October 2006.\par A(i,j) is the edge from node i to node j in the graph. If a graph is not\par listed as weighted, and yet has non-binary entries, then the entry a(i,j)\par reflects the number of edges (i,j) in the original data (the graph is a\par multigraph). In some cases, it was clear that the extra entries were\par duplicates, and the graph should not be a multigraph (problems GD97, EAT_RS,\par and patents).\par \par All data was converted without loss of information, except where intentional.\par Details are given below.\par \par The largest graph is "patents", with over 3.7 million nodes and nearly\par 15 million entries.\par \par "Pajek" is Slovenian for "spider" and is roughly pronounced "Pie yeck".\par \par The Pajek networks are all 1-based (nodes are numbered starting at node 1).\par Some networks when included in the Pajek Dataset were converted from 0-based\par to 1-based. This was done by renaming node 0 as a new node n, and leaving\par all other nodes unchanged. This differs from the MATLAB convention, which\par is to add 1 to all row/column indices. This conversion was left as-is\par (networks: USpowerGrid, EPA, and Kleinberg) when included in the UF\par Sparse Matrix Collection.\par \par Remember that in MATLAB, A(i,:) is slow to compute; A(:,i) is much faster. If\par you want row i of a sparse matrix, access the ith column of the transpose\par instead.\par\par \par ================================================================================\par Summary:\par ============================== http://www.downhi.com/txt/PMhieofiFTRK.html ==================================================\par \par CSphd: PhD's in computer science\par 1882-by-1882 with 1740 nonzeros\par kind: directed graph with auxiliary node data\par \par California: Kleinberg's web search of "California"\par 9664-by-9664 with 16150 nonzeros\par kind: directed graph\par \par Cities: www.lboro.ac.uk/gawc, data set 6\par 55-by-46 with 1342 nonzeros\par kind: weighted bipartite graph with auxiliary node data\par \par EAT_RS: Edinburgh Associative Thesaurus (response-stimulus)\par 23219-by-23219 with 325592 nonzeros\par kind: directed weighted graph\par \par EAT_SR: Edinburgh Associative Thesaurus (stimulus-response)\par 23219-by-23219 with 325589 nonzeros\par kind: directed weighted graph\par \par EPA: Kleinberg, pages linking to www.epa.gov\par 4772-by-4772 with 8965 nonzeros\par kind: directed graph\par \par EVA: EVA, corporate inter-relationships\par 8497-by-8497 with 6726 nonzeros\par kind: directed graph\par \par Erdos02: Erdos collaboration network\par 6927-by-6927 with 16944 nonzeros\par kind: undirected graph\par \par Erdos971: Erdos collaboration network\par 472-by-472 with 2628 nonzeros\par kind: undirected graph\par \par Erdos972: Erdos collaboration network\par 5488-by-5488 with 14170 nonzeros\par kind: undirected graph with auxiliary node data\par \par Erdos981: Erdos collaboration network\par 485-by-485 with 2762 nonzeros\par kind: undirected graph\par \par Erdos982: Erdos collaboration network\par 5822-by-5822 with 14750 nonzeros\par kind: undirected graph with auxiliary node data\par \par Erdos991: Erdos collaboration network\par 492-by-492 with 2834 nonzeros\par kind: undirected graph\par \par Erdos992: Erdos collaboration network\par 6100-by-6100 with 15030 nonzeros\par kind: undirected graph with auxiliary node data\par \par FA: USF Free (word) Associa http://www.downhi.com/txt/PMhieofiFTRK.html tion Norms\par 10617-by-10617 with 72176 nonzeros\par kind: directed weighted graph with auxiliary node data\par \par GD00_a: Graph Drawing contest 2000\par 352-by-352 with 458 nonzeros\par kind: directed graph\par \par GD00_c: Graph Drawing contest 2000\par 638-by-638 with 1041 nonzeros\par kind: directed multigraph\par \par GD01_A: Graph Drawing contest 2001\par 953-by-953 with 645 nonzeros\par kind: directed multigraph\par \par GD01_a: Graph Drawing contest 2000\par 311-by-311 with 645 nonzeros\par kind: directed weighted graph\par \par GD01_b: Graph Drawing contest 2001\par 18-by-18 with 37 nonzeros\par kind: directed graph\par \par GD01_c: Graph Drawing contest 2001\par 33-by-33 with 135 nonzeros\par kind: directed multigraph with auxiliary node data\par \par GD02_a: Graph Drawing contest 2002\par 23-by-23 with 87 nonzeros\par kind: directed graph\par \par GD02_b: Graph Drawing contest 2002\par 80-by-80 with 232 nonzeros\par kind: directed graph\par \par GD06_Java: Graph Drawing contest 2006\par 1538-by-1538 with 8032 nonzeros\par kind: directed graph\par \par GD06_theory: Graph Drawing contest 2006\par 101-by-101 with 380 nonzeros\par kind: undirected graph\par \par GD95_a: Graph Drawing contest 1995\par 36-by-36 with 57 nonzeros\par kind: directed graph\par \par GD95_b: Graph Drawing contest 1995\par 73-by-73 with 96 nonzeros\par kind: directed graph\par \par GD95_c: Graph Drawing contest 1995\par 62-by-62 with 287 nonzeros\par kind: directed graph\par \par GD96_a: Graph Drawing contest 1996\par 1096-by-1096 with 1677 nonzeros\par kind: directed multigraph\par \par GD96_b: Graph Drawing contest 1996\par 111-by-111 with 193 nonzeros\par kind: directed graph\par \par GD96_c: Graph Drawing contest 1996\par 65-by-65 with 250 nonzeros\par kind: undirected graph\par \par http://www.downhi.com/txt/PMhieofiFTRK.html GD96_d: Graph Drawing contest 1996\par 180-by-180 with 229 nonzeros\par kind: directed graph\par \par GD97_a: Graph Drawing contest 1997\par 84-by-84 with 332 nonzeros\par kind: directed graph\par \par GD97_b: Graph Drawing contest 1997\par 47-by-47 with 264 nonzeros\par kind: undirected weighted graph\par \par GD97_c: Graph Drawing contest 1997\par 452-by-452 with 460 nonzeros\par kind: directed multigraph\par \par GD98_a: Graph Drawing contest 1998\par 38-by-38 with 50 nonzeros\par kind: directed graph\par \par GD98_b: Graph Drawing contest 1998\par 121-by-121 with 207 nonzeros\par kind: directed graph\par \par GD98_c: Graph Drawing contest 1998\par 112-by-112 with 336 nonzeros\par kind: undirected graph\par \par GD99_b: Graph Drawing contest 1999\par 64-by-64 with 252 nonzeros\par kind: undirected multigraph\par \par GD99_c: Graph Drawing contest 1999\par 105-by-105 with 149 nonzeros\par kind: directed graph with auxiliary node data\par \par GlossGT: graph and digraph glossary\par 72-by-72 with 122 nonzeros\par kind: directed graph\par \par HEP-th-new: High Energy Physics literature\par 27770-by-27770 with 352807 nonzeros\par kind: directed graph with auxiliary node data\par \par HEP-th: High Energy Physics literature\par 27240-by-27240 with 342437 nonzeros\par kind: directed graph\par \par IMDB: IMDB movie/actor network, www.imdb.com\par 428440-by-896308 with 3782463 nonzeros\par kind: bipartite graph with auxiliary node data\par \par Journals: Slovenian journals 1999-2000\par 124-by-124 with 12068 nonzeros\par kind: undirected weighted graph with auxiliary node data\par \par Kohonen: Kohonen citation network\par 4470-by-4470 with 12731 nonzeros\par kind: directed graph with auxiliary node data\par \par Lederberg: Lederberg citation network\par 8843-by-8843 with 41601 nonzeros\par http://www.downhi.com/txt/PMhieofiFTRK.html kind: directed multigraph with auxiliary node data\par \par NotreDame_actors: Barabasi's actor network (of www.imdb.com)\par 392400-by-127823 with 1470404 nonzeros\par kind: bipartite multigraph\par \par NotreDame_www: Barabasi's web page network of nd.edu\par 325729-by-325729 with 929849 nonzeros\par kind: directed graph\par \par NotreDame_yeast: Barabasi's yeast protein interaction\par 2114-by-2114 with 4480 nonzeros\par kind: undirected graph\par \par ODLIS: online dictionary of library & inf. sci\par 2909-by-2909 with 18246 nonzeros\par kind: directed multigraph\par \par Ragusa16: Ragusa set\par 24-by-24 with 81 nonzeros\par kind: directed weighted graph\par \par Ragusa18: Ragusa set\par 23-by-23 with 64 nonzeros\par kind: directed weighted graph\par \par Reuters911: Reuters news, Sept 11 to Nov 15, 2001\par 13332-by-13332 with 296076 nonzeros\par kind: temporal undirected weighted graph\par \par Roget: Roget's Thesaurus, 1879\par 1022-by-1022 with 5075 nonzeros\par kind: directed graph\par \par Sandi_authors: Klavzar bibliography\par 86-by-86 with 248 nonzeros\par kind: undirected weighted graph\par \par Sandi_sandi: Klavzar bibliography\par 314-by-360 with 613 nonzeros\par kind: bipartite graph\par \par SciMet: SciMet citation network\par 3084-by-3084 with 10413 nonzeros\par kind: directed multigraph with auxiliary node data\par \par SmaGri: SmaGri citation network\par 1059-by-1059 with 4919 nonzeros\par kind: directed multigraph with auxiliary node data\par \par SmallW: SmallW citation network\par 396-by-396 with 994 nonzeros\par kind: directed multigraph with auxiliary node data\par \par Stranke94: Slovene Parliamentary Parties 1994\par 10-by-10 with 90 nonzeros\par kind: undirected weighted graph\par \par Tina_AskCal: student govt, Univ. Ljubljana, 1992 (ask opin., recall)\par 11-by-11 with http://www.downhi.com/txt/PMhieofiFTRK.html 29 nonzeros\par kind: directed graph\par \par Tina_AskCog: student govt, Univ. Ljubljana, 1992 (ask, recognized)\par 11-by-11 with 36 nonzeros\par kind: directed graph\par \par Tina_DisCal: student govt, Univ. Ljubljana, 1992 (discuss, recall)\par 11-by-11 with 41 nonzeros\par kind: directed graph\par \par Tina_DisCog: student govt, Univ. Ljubljana, 1992 (discuss, recog.)\par 11-by-11 with 48 nonzeros\par kind: directed graph\par \par USAir97: US Air flights, 1997\par 332-by-332 with 4252 nonzeros\par kind: undirected weighted graph\par \par USpowerGrid: US power grid\par 4941-by-4941 with 13188 nonzeros\par kind: undirected multigraph\par \par Wordnet3: Wordnet3 dictionary network\par 82670-by-82670 with 132964 nonzeros\par kind: directed weighted graph with auxiliary node data\par \par WorldCities: world city network\par 315-by-100 with 7518 nonzeros\par kind: weighted bipartite graph with auxiliary node data\par \par Zewail: Zewail citation network\par 6752-by-6752 with 54233 nonzeros\par kind: directed multigraph with auxiliary node data\par \par dictionary28: dictionary\par 52652-by-52652 with 178076 nonzeros\par kind: undirected graph\par \par divorce: divorce laws in the 50 US states\par 50-by-9 with 225 nonzeros\par kind: bipartite graph\par \par foldoc: free on-line dictionary of computing\par 13356-by-13356 with 120238 nonzeros\par kind: directed weighted graph\par \par football: World Soccer, Paris 1998\par 35-by-35 with 118 nonzeros\par kind: directed weighted graph\par \par geom: collaboration in computational geometry\par 7343-by-7343 with 23796 nonzeros\par kind: undirected weighted graph\par \par internet: connectivity of internet routers\par 124651-by-124651 with 207214 nonzeros\par kind: directed weighted graph\par \par patents: NBER US Patent Citations, 1963-1999, cites 1975-1999\par http://www.downhi.com/txt/PMhieofiFTRK.html 3774768-by-3774768 with 14970767 nonzeros\par kind: directed graph with auxiliary node data\par \par patents_main: main NBER US Patent Citations, 1963-1999, cites 1975-1999\par 240547-by-240547 with 560943 nonzeros\par kind: directed weighted graph with auxiliary node data\par \par yeast: yeast protein interaction network\par 2361-by-2361 with 13828 nonzeros\par kind: undirected graph with auxiliary node data\par\par \par \par ================================================================================\par ==> Citations/Cite2001.txt <==\par ================================================================================\par \par Citation networks\par \par Graph Drawing 2001 Contest - Graph A Graph Drawing 2001\par \par Contest task description for Graph A. Graph A in Pajek's format. Selected\par citation networks from Garfield's collection Citation networks in Pajek's\par format obtained from the Garfield's collection of citation network datasets\par produced using HistCite software. All of these networks are the result of\par searches in the WebofScience and are used with the permission of ISI of\par Philadelphia. Please acknowledge this when publishing results based on these\par data.\par \par SmallW: Papers that cite S Milgram's 1967 Psychology Today paper or use\par Small World in title, Tue Jul 23 13:35:11 2002\par \par SmaGri: Citations to Small & Griffith and Descendants, Thu Nov 8 10:40:55\par 2001 \par \par SciMet: Articles from or citing Scientometrics, 1978-2000, Wed Jun\par 12 16:39:51 2002 \par \par Kohonen: Articles with topic "self-organizing maps"\par or references to "Kohonen T", Tue Jun 18 10:39:51 2002\par \par Zewail: Articles citing and by AH Zewail, 1970-2002, Wed Jul 31 15:46:38\par 2002 \par \par Lederberg: Articles by and citing J Lederberg, 1945-2002, Wed Jul 31\par 1 http://www.downhi.com/txt/PMhieofiFTRK.html 3:40:22 2002 \par \par Some references\par \par * Batagelj V.: Some Mathematics of Network Analysis. Network Seminar,\par Department of Sociology, University of Pittsburgh, January 21, 1991.\par \par * Batagelj V., Mrvar A.: Graph Drawing Contest 2001 Layouts.\par \par * Garfield E, Sher IH, and Torpie RJ.: The Use of Citation Data in Writing\par the History of Science. Philadelphia: The Institute for Scientific\par Information, December 1964.\par \par * Garfield E.: From Computational Linguistics to Algorithmic\par Historiography, paper presented at the Symposium in Honor of Casimir\par Borkowski at the University of Pittsburgh School of Information Sciences,\par September 19, 2001.\par \par * Hummon N.P., Doreian P.: Connectivity in a Citation Network: The\par Development of DNA Theory. Social Networks, 11(1989) 39-63.\par \par Network Data, Pajek, Vlado\par\par \par ================================================================================\par ==> Cities/Url.txt <==\par ================================================================================\par \par http://www.lboro.ac.uk/gawc/data.html\par \par World City Relation Data\par \par This is the raison d'�tre of GaWC. For an introduction, see A Brief Guide to\par Quantitative Data Collection at GaWC, 1997-2001.\par \par INVITATION\par \par World city researchers with relational data on world cities are encouraged to\par post it here. Depositors may apply their own protocol for use by others.\par (Contact p.j.taylor@lboro.ac.uk) PROTOCOL for using GaWC data\par \par As part of our mission to promote the study of non-state relations we wish to\par encourage people to use the data posted here in their research and teaching.\par However, we do ask that, as a matter of courtesy bordering on ethics, the\par research project and the information gatherers a http://www.downhi.com/txt/PMhieofiFTRK.html re properly acknowledged along\par with GaWC itself. Each data set will have a statement as to whom to acknowledge\par and we ask that this be reproduced in every public use of the data.\par\par \par \par Data Set 1: US Cities: Surrogate Measures of Relations, 1990 (T.R. Longcore, C.\par McWilliams and P.J. Taylor)\par \par Data Set 2: London and New York: Surrogate Measures of Relations, 1997\par \par Data Set 3: Randstad Cities: Surrogate Measures of Relations, 1970-95 (A.M.\par Beerda)\par \par Data Set 4: London's Relations with other Cities Using Producer Service Office\par Geographies (J.V. Beaverstock, R.G. Smith and P.J. Taylor)\par \par Data Set 5: US Cities: Law Firm Office Geographies, 1998 (J.V. Beaverstock,\par R.G. Smith and P.J. Taylor)\par \par Data Set 6: World Cities and Global Firms (P.J. Taylor and D.R.F. Walker)\par \par Data Set 7: Inter-City Matrices (P.J. Taylor and D.R.F. Walker)\par \par Data Set 8: World Cities: Regional Dimensions (P.J. Taylor, D.R.F. Walker and\par M. Hoyler)\par \par Data Set 9: Cities Mentioned in Advertisements in The Economist (May 2000 -\par January 2001) (P.J. Taylor)\par \par Data Set 10: The Relative Centrality of Cities Based upon Air Passenger Travel,\par 1977-1997 (M. Timberlake and D.A. Smith with K.-H. Shin)\par \par Data Set 11: World City Network: The Basic Data (P.J. Taylor and G. Catalano)\par \par Data Set 12: Global Network Service Connectivities for 315 Cities in 2000 (P.J.\par Taylor)\par\par \par DATA TOOLS\par \par Data Tool 1: Macro for Calculating Connectivities (max 254 cities x 255 firms)\par (E.C. Rossi and C.C.C. Rossi)\par \par Data Tool 2: Macro for Calculating Connectivities (max 1100 cities x 255\par firms) (R. Aranya)\par\par \par \par World Cities and Global Firms\par \par P.J. Taylor and D.R.F. Walker\par \par These data consist of the distribution of offices for 46 'global' advanced http://www.downhi.com/txt/PMhieofiFTRK.html \par producer service firms over 55 world cities. Global firms are defined by having\par offices in at least 15 different cities. World cities are from the GaWC\par inventory of world cities (see GaWC Research Bulletin 6). Service values for a\par firm in a city are given as 3, 2, 1 or 0 as defined in Data Set 4.\par \par These data are an experimental set of data derived from Data Set 4 (43 of the\par firms qualify as global) but with three additional law firms added which do not\par have London offices. For publications that make use of these data, see GaWC\par Research Bulletin 13 and GaWC Research Bulletin 17.\par\par \par Key to Data Set 6:\par \par ADVANCED PRODUCER SERVICE FIRMS (SECTOR, TABLE CODE AND FIRM)\par \par Accountancy\par \par KP KPMG\par CL Coopers & Lybrand\par EY Ernst & Young International\par AA Arthur Andersen\par PW Price Waterhouse\par \par Advertising\par \par GR Grey Worldwide\par DM DMB&B (MacManus Group)\par LH Lowe Howard -Spink\par SS Saatchi and Saatchi\par TH JWT (Thompson)\par OM Ogilvy & Mather Direct Worldwide\par DE Dentsu\par YR Young & Rubicam\par TM TMP\par PU Publicis\par AM Abbott Mead Vickers (BBDO)\par \par Banking and Finance\par \par HS HSBC\par BA Barclays\par NW NatWest Group\par SC Standard Chartered Group\par CR Creditanstalt-Bankverein\par DR Dresdner Bank Group\par MO J P Morgan\par PA Compagnie Financi�re de Paribas SA\par CS Credit Suisse\par BB BBV Group\par BT Banker's Trust\par UB UBS\par AB ABN-AMRO\par CB Citibank\par \par Law\par \par BM Baker & McKenzie\par WC White & Case\par LL Leboeuf, Lamb, Greene & MacRae\par CO Coudert Brothers\par SK Skadden Arps\par BC Bryan Cave\par DW Dorsey & Whitney\par GJ Graham & James\par HH Hogan & Hartson\par JD Jones, Day, Reavis & Pogue\par MC Miller, Canfield, Paddock & Stone\par SQ Squire, Sanders & Dempsey\par WE Wilson, Elser, Moskowitz, Edelman & Dicker http://www.downhi.com/txt/PMhieofiFTRK.html \par AO Allen & Overy\par CC Clifford Chance\par FF Freshfields\par \par As per our data protocol, the following acknowledgement should accompany any\par public use of the data:\par \par ACKNOWLEDGEMENT: The data used is from Data Set 6 from the GaWC Study Group and\par Network (http://www.lboro.ac.uk/gawc/). It was created by P.J. Taylor and\par D.R.F. Walker as part of their project "World City Network: Data Matrix\par Construction and Analysis" and is based on primary data collected by J.V.\par Beaverstock, R.G. Smith and P.J. Taylor (ESRC project "The Geographical Scope\par of London as a World City" (R000222050)).\par\par \par ================================================================================\par ==> Days/Days.txt <==\par ================================================================================\par \par Pajek datasets\par \par Reuters terror news network (NOTE: renamed Reuters911 for UF sparse collection)\par \par Dataset days\par \par Description\par \par days.net undirected temporal network with 13332 vertices and 243447 edges.\par\par \par \par Days, DaysAll. Background\par \par Reuters terror news network Days.net in Pajek's format obtained from the CRA\par networks produced by Steve Corman and Kevin Dooley at Arizona State University.\par Please acknowledge this when publishing results based on these data.\par \par The Reuters terror news network is based on all stories released during 66\par consecutive days by the news agency Reuters concerning the September 11 attack\par on the U.S., beginning at 9:00 AM EST 9/11/01. The vertices of a network are\par words (terms); there is an edge between two words iff they appear in the same\par text unit (sentence). The weight of an edge is its frequency. The network has n\par = 13332 vertices (different words in the news) and m = 243447 edges, 50859 with\par value larger than 1. There are no loops in the netw http://www.downhi.com/txt/PMhieofiFTRK.html ork.\par \par The network DaysAll.net contains the main connected component of the network\par obtained by transforming the Reuters terror news network into a combined\par network for all 66 days (union of all time points). It has 13308 vertices.\par \par The Reuters terror news network was used as a case network for the Viszards\par visualization session on the Sunbelt XXII International Sunbelt Social Network\par Conference, New Orleans, USA, 13-17. February 2002. \par \par History\par \par 1. 5-12. Dec 2001, networks constructed by S. Corman and his group\par \par 2. December 2001: CRA data transformed in Pajek format by V. Batagelj.\par \par References\par \par 1. Steven R. Corman, Timothy Kuhn, Robert D. Mcphee and Kevin J. Dooley\par (2002): Studying Complex Discursive Systems: Centering Resonance Analysis of\par Communication. (PDF)\par \par 2. Crawdad Technologies\par \par 3. Batagelj, V., & Mrvar, A. (2003): A density based approaches to network\par analysis: Analysis of Reuters terror news network, Ninth Annual ACM SIGKDD,\par Washington, D.C. (PDF); ( SVG)\par \par 4. Jeffrey C. Johnson and Lothar Krempel (2004): Network Visualization: The\par "Bush Team" in Reuters News Ticker 9/11-11/15/01. The Journal of Social\par Structure's, Vol. 5, No. 1. ( HTML)\par \par Pajek Data; 20. April 2006 / 27. January 2004\par \par -------------------------------------------------------------------------------\par When converted to a sparse adjacency matrix for the UF Sparse Matrix\par Collection, Day{i} is the graph of the ith day. The diagonal entry\par Day{i}(k,k) is 1 if word k appears in any news on the ith day. Note\par that it may not appear in conjunction with other words in the same\par sentence on that day. The sum of nnz(tril(Day{i})) for i=1:66 is 243,447.\par The overall matrix A is the sum of the Day{i} matrices. A(i,j) is the n http://www.downhi.com/txt/PMhieofiFTRK.html umber\par of times words i and j appear in same sentence (for i not equal to j). A(k,k)\par is the number of days the word k appears in any news report.\par Note that this network has been renamed to Reuters911 here.\par\par \par ================================================================================\par ==> EAT/EAT.txt <==\par ================================================================================\par\par \par Pajek datasets EAT The Edinburgh Associative Thesaurus\par \par Dataset eat\par \par Description\par \par eatRS.net directed network with 23219 vertices and 325624 arcs (564 loops);\par stimulus X is associated with response Y N times. eatSR.net directed network\par with 23219 vertices and 325589 arcs (564 loops); response X is associated with\par stimulus Y N times.\par \par It seems that the SR network is incomplete and that it should be the inverse of\par RS network.\par\par \par \par EAT response-stimulus (ZIP, 1321K) EAT stimulus-response (ZIP, 1306K)\par Background\par \par The Edinburgh Associative Thesaurus (EAT) is a set of word association norms\par showing the counts of word association as collected from subjects. This is not\par a developed semantic network such as WordNet (3), but empirical association\par data.\par \par The traditional way to collect word association norms is to show or say a word\par to several people and ask them to say the word which first comes to their minds\par upon receiving the stimulus. The link established between the stimulus and the\par response is not semantically labelled (e.g. as synonym, antonym or by a case\par relation) and can only be regarded as an association.\par \par The Edinburgh association norms were collected by growing the network from a\par nucleus set of words. Responses were collected to words in this nucleus set,\par then these responses were used to obtain further responses, and so on. In fac http://www.downhi.com/txt/PMhieofiFTRK.html t\par the cycle was repeated about three times since by then the number of different\par responses was so large that they could not be re-used as stimuli. Data\par collection stopped when 8400 stimulus words had been used. Each stimulus word\par was presented to 100 different subjects, each of whom received 100 words. This\par gave rise to a total of 55732 nodes in the Thesaurus network.\par \par The subjects were mostly undergraduates from a wide variety of British\par universities. The age range of the subjects was from 17 to 22 with a mode of\par 19. The sex distribution was 64 per cent male and 36 per cent female. The data\par was collected between June 1968 and May 1971.\par \par The database consists of two files. The SR (stimulus-response) file, and the RS\par (response-stimulus) file. Where words have been truncated to 19 characters to\par save space the per cent character (%) has been placed as the 20th.\par \par The EAT here is that included in the MRC Psycholinguistic Database (4), for use\par with the other measures available there. EAT Data Collection Procedure (1)\par Stimulus words\par \par Since the objective was to obtain a reasonably large complete mapping of the\par associative network for a large set of words, a systematic procedure of\par 'growing' the network from a small nucleus was followed. At first responses\par were obtained from this nucleus set, then these responses were used as stimuli\par to obtain further responses, and so on. In fact, this cycle was repeated about\par three times, since by then the number of different responses was so large that\par they could not all be re-used as stimuli.\par \par The nucleus set was derived from (a) the 200 stimuli used in the Palermo and\par Jenkins (1964) normq (b) the 1,000 most frequent words of the Thorndike and\par Lorge (1944) word frequency count and (c) the basic English vocabulary of Ogden\par (1954).\par \pa http://www.downhi.com/txt/PMhieofiFTRK.html r Data collection was stopped when 8,400 stimulus words had been used. Only a\par minimal amount of selection of stimuli was applied in each cycle of the data\par collection. Effectively all responses which were English words or meaningful\par verbal units were included, including some phrasal forms and numerals. The data\par cover a wide range of grammatical form classes and inflexional forms.\par Procedure\par \par Each stimulus word was presented to 100 different subjects. Each subject\par recieved a computer-printed sheet with 100 stimuli in randomised arrangement\par (to minimize priming effects). The total contribution of each subject was thus\par 100 responses. The verbal environment of each word for each subject was\par different. The instructions asked the subject to write down against each\par stimulus the first word it made him think of, working as quickly as possible.\par the total time spent on this task was measured, and most subjects completed the\par sheet in five to ten minutes.\par \par Most of the data was collected in a classroom setting under supervision. Sheets\par which had more than 25 percent blank responses were rejected and fresh data was\par collected.\par \par History\par \par 1. Original EAT: George Kiss, Christine Armstrong, Robert Milroy and J.R.I.\par Piper (collected between June 1968 and May 1971).\par \par 2. MRC Psycholinguistic Database Version modified by: Max Coltheart, S.\par James, J. Ramshaw, B.M. Philip, B. Reid, J. Benyon-Tinker and E. Doctor;\par made available by: Philip Quinlan.\par \par 3. The present version was re-structured and documented by Michael Wilson at\par the Rutherford Appleton Laboratory in 1988 (2).\par \par 4. transformed in Pajek format: V. Batagelj, 31. July 2003.\par\par \par References\par \par 1. Kiss, G.R., Armstrong, C., Milroy, R., and Piper, J. (1973) An\par associative thesaurus of http://www.downhi.com/txt/PMhieofiFTRK.html English and its computer analysis. In Aitken, A.J.,\par Bailey, R.W. and Hamilton-Smith, N. (Eds.), The Computer and Literary\par Studies. Edinburgh: University Press.\par \par 2. The present version of The Edinburgh Associative Thesaurus (ZIP, 2.7M)\par \par 3. WordNet\par \par 4. MRC Psycholinguistic Database\par \par 5. Coltheart, M. (1981) MRC Psycholinguistic Database. Quarterly Journal of\par Experimental Psychology, 3A, 497-505.\par \par 6. MRC Psycholinguistic Database 2\par \par Pajek Data; 31. July 2003\par \par -------------------------------------------------------------------------------\par NOTE regarding conversion for UF sparse matrix collection: in the original data\par there are 325,624 weighted edges. Of those only 32 edges are duplicates, and \par all of them have identical edge weights as the edges they are duplicates of \par These extraneous edges have been removed, since this this appears to be a \par graph, not a multigraph. \par\par \par ================================================================================\par ==> EVA/EVA.txt <==\par ================================================================================\par \par Pajek datasets\par \par EVA Extraction, Visualization & Analysis of corporate\par inter-relationships\par \par Dataset EVA\par \par Description\par \par EVA.net directed network with 8343 vertices and 6726 arcs.\par\par \par \par EVA.net (ZIP, 204K); included also original files names.txt and ownership.txt.\par Background\par \par EVA / Denali is a multidisciplinary research project combining information\par extraction, information visualization, and social network analysis techniques\par to bring greater transparency to the public disclosure of inter-relationships\par between corporations. The project is described in the paper [1].\p http://www.downhi.com/txt/PMhieofiFTRK.html ar \par Abstract: We present EVA, a prototype system for extracting, visualizing, and\par analyzing corporate ownership information as a social network. Using\par probabilistic information retrieval and extraction techniques, we automatically\par extract ownership relationships from heterogeneous sources of online text,\par including corporate annual reports (10-Ks) filed with the U.S. Securities and\par Exchange Commission (SEC). A browser-based visualization interface allows users\par to query the relationship database and explore large networks of companies.\par Applying the system and methodology to the telecommunications and media\par industries, we construct an ownership network with 6,726 relationships among\par 8,343 companies. Analysis reveals a highly clustered network, with over 50% of\par all companies connected to one another in a single component. Furthermore,\par ownership activity is highly skewed: 90% of companies are involved in no more\par than one relationship, but the top ten companies are parents for over 24% of\par all relationships. We are also able to identify the most influential companies\par in the network using social network analysis metrics such as degree,\par betweenness, cutpoints, and cliques. We believe this methodology and tool can\par aid government regulators, policy researchers, and the general public to\par interpret complex corporate ownership structures, thereby bringing greater\par transparency to the public disclosure of corporate inter-relationships.\par \par Note that we do not have ownership relationships for all companies, so there\par will be companies without links.\par \par An arc (X,Y) from company X to company Y exists in the network if in the\par company X is an owner of company Y.\par \par Copyright 2002 by Denali Project. If you use this dataset in your research,\par please use the citation to paper [1] as the source of the data.\par \ http://www.downhi.com/txt/PMhieofiFTRK.html par "Denali" is the Native American name for the tallest peak in North America. It\par means "the Great One."\par \par If you have any questions, please contact: John Chuang, Mike Gebbie, Gabe\par Lucas, Kim Norlen. \par \par History\par \par 1. 2002 collection of original data by the EVA group; 2. March 6, 2004:\par original data transformed into Pajek format EVA.net by V. Batagelj.\par \par References\par \par 1. Kim Norlen, Gabriel Lucas, Mike Gebbie, and John Chuang. EVA: Extraction,\par Visualization and Analysis of the Telecommunications and Media Ownership\par Network. Proceedings of International Telecommunications Society 14th\par Biennial Conference, Seoul Korea, August 2002. (paper berkeley / local;\par slides berkeley / local)\par \par Pajek Data; 6. March 2004\par\par \par ================================================================================\par ==> Foldoc/Foldoc.txt <==\par ================================================================================\par \par Pajek datasets FOLDOC Free On-line Dictionary of Computing\par \par Dataset Foldoc\par \par Description\par \par foldoc.net valued directed network with 13380 vertices and 120700 arcs, value\par is the multiplicity of arc.\par\par \par \par foldoc.net (ZIP, 517K) Background\par \par FOLDOC is a searchable dictionary of acronyms, jargon, programming languages,\par tools, architecture, operating systems, networking, theory, conventions,\par standards, mathematics, telecoms, electronics, institutions, companies,\par projects, products, history, in fact anything to do with computing.\par \par The dictionary has been growing since 1985 and now contains over 13000\par definitions totalling nearly five megabytes of text. Entries are\par cross-referenced to each other and to related resources elsewhere on the net.\par \par An arc (X,Y) from term X to term Y exists in the network http://www.downhi.com/txt/PMhieofiFTRK.html iff in the FOLDOC\par dictionary the term Y is used to describe the meaning of term X.\par \par Copyright 1993 by Denis Howe. Permission is granted to copy, distribute and/or\par modify the FOLDOC dictionary under the terms of the GNU Free Documentation\par License, Version 1.1 or any later version published by the Free Software\par Foundation.\par \par Please refer to the dictionary as "The Free On-line Dictionary of Computing,\par http://www.foldoc.org/, Editor Denis Howe" or similar. \par \par History\par \par 1. FOLDOC started in 1985 by Denis Howe;\par \par 2. in 1990 put on FTP, other sources included;\par \par 3. from 1994 available on the WWW, hundreds of contributors;\par \par 4. February/June 2002: Foldoc.net transformed in Pajek format and 'cleaned'\par by V. Batagelj and A. Mrvar.\par \par References\par \par 1. Denis Howe, Editor: FOLDOC (2002): Free on-line dictionary of computing.\par \par 2. V. Batagelj, A. Mrvar, M. Zavešnik: Network analysis of texts. Language\par Technologies 2002, Ljubljana, 14 - 15th October 2002, p. 143-148.\par \par 3. V. Batagelj, A. Mrvar, M. Zavešnik: Network analysis of dictionaries.\par Language Technologies 2002, Ljubljana, 14 - 15th October 2002, p. 135-142.\par\par \par Pajek Data; 30. January 2004\par\par \par ================================================================================\par ==> Football/Football.txt <==\par ================================================================================\par \par Pajek datasets World Soccer Data Paris 1998\par \par Dataset Football\par \par Description\par \par football.net - valued network with 35 vertices\par\par \par \par complete dataset\par \par Background\par \par Our network example describes the 22 soccer teams which participated in the\par World Championship in Paris, 1998.\par \par Players of the national team often have contracts in ot http://www.downhi.com/txt/PMhieofiFTRK.html her countries. This\par constitutes a players market where national teams export players to other\par countries. Members of the 22 teams had contracts in altogether 35 countries.\par \par Counting which team exports how many players to which country can be described\par with a valued, asymmetric graph. The graph is highly unsymmetric: some\par countries only export players, some countries are only importers. \par \par History\par \par 1. Data collected by Lothar Krempel, October 5, 1999\par \par 2. Transformed in Pajek format by V. Batagelj, February 9, 2001.\par \par References\par \par 1. Dagstuhl seminar: Link Analysis and Visualization, Dagstuhl 1-6. July\par 2001\par \par Pajek Data; 21. September 2005\par\par \par ================================================================================\par ==> Geom/Geom.txt <==\par ================================================================================\par\par \par Pajek datasets Geom Collaboration network in computational geometry\par \par Dataset Geom\par \par Description\par \par Geom.net valued undirected network with 7343 vertices and 11898 edges; author X\par wrote a joint work with author Y; value is the number of joint works.\par\par \par \par Geom.net (ZIP, 139K)\par \par Background\par \par The network Geom.net is based on the file geombib.bib that contains\par Computational Geometry Database, version February 2002.\par \par The authors collaboration network in computational geometry was produced from\par the BibTeX bibliography [Beebe, 2002] obtained from the Computational Geometry\par Database geombib, version February 2002 [Jones, 2002].\par \par Two authors are linked with an edge, iff they wrote a common work (paper, book,\par ...). The value of an edge is the number of common works. Using a simple\par program written in programming language Python, the BibTeX data were\par transformed into th http://www.downhi.com/txt/PMhieofiFTRK.html e corresponding network, and output to the file in Pajek\par format.\par \par The obtained network has 9072 vertices (authors) and 22577 edges (common papers\par or books) / 13567 edges as a simple network - multiple edges between a pair of\par authors are replaced with a single edge.\par \par The problem with the obtained network is that, because of non standardized\par writing of the author's name, it contains several vertices corresponding to the\par same author. For example:\par \par R.S. Drysdale, Robert L. Drysdale, Robert L. Scot Drysdale, R.L. Drysdale,\par S. Drysdale, R. Drysdale, and R.L.S. Drysdale;\par \par or:\par \par Pankaj K. Agarwal, P. Agarwal, Pankaj Agarwal, and P.K. Agarwal\par \par that are easy to guess; but an 'insider' information is needed to know that\par Otfried Schwarzkopf and Otfried Cheong are the same person. Also, no provision\par is made in the database to discern two persons with the same name. We manually\par produced the name equivalence partition and then shrank (in Pajek) the network\par according to it.\par \par The reduced simple network contains 7343 vertices and 11898 edges. It is a\par sparse network - its average degree is 2m/n = 3.24.\par \par History\par \par 1. Computational Geometry Database started in 1986 by merging two lists of\par references - one compiled by Edelsbrunner and van Leeuwen and the other by\par Guibas and Stolfi;\par \par 2. Computational Geometry Database, February 2002 Edition;\par \par 3. March-April 2002: Geom.bib transformed in Pajek format and 'cleaned' by\par V. Batagelj and M. Zaveršnik.\par \par References\par \par 1. Beebe, N.H.F. (2002): Nelson H.F. Beebe's Bibliographies Page.\par \par 2. Jones, B., Computational Geometry Database, February 2002; FTP / HTTP\par \par Pajek Data; 27. January 2004\par\par \par ==================================================================== http://www.downhi.com/txt/PMhieofiFTRK.html ============\par ==> GlossGT/GlossGT.txt <==\par ================================================================================\par \par Pajek datasets Graph and Digraph Glossary\par \par Dataset glossGT\par \par Description\par \par glossGT.net mixed network with 72 vertices and 114 arcs and 4 edges; word X is\par related to word Y.\par\par \par \par glossTG.paj (4K)\par \par Background\par \par The network GlossGT.net is based on the file glossary.html containing Bill\par Cherowitzo's Graph and Digraph Glossary. An arc (X,Y) from term X to term Y\par exists in the network iff in the Graph and Digraph Glossary the term Y is used\par to describe the meaning of term X.\par \par History\par \par 1. 1998-2001, Bill Cherowitzo prepared the glossary. 2. Graph and Digraph\par Glossary transformed in Pajek format: Barbara Zemlji"c, 2. nov 2003.\par \par References\par \par 1. Bill Cherowitzo: Graph and Digraph Glossary, version 03-Feb-2001; (Course\par page)\par \par Pajek Data; 25. January 2004\par\par \par ================================================================================\par ==> HEP-th/HEP.txt <==\par ================================================================================\par\par \par Pajek datasets KDD Cup 2003 High Energy Particle Physics (HEP) literature\par \par Dataset hep-th\par \par Description\par \par hep-th.net directed network with 27240 vertices and 342437 arcs (39 loops).\par hep-th-new.net directed network with 27770 vertices and 352807 arcs (39 loops).\par date-new.vec integer vector on 27770 vertices. year-new.vec integer vector on\par 27770 vertices.\par\par \par \par complete dataset (ZIP, 2607K)\par \par Background Citation data from KDD Cup 2003, a knowledge discovery and data\par mining competition held in conjunction with the Ninth Annual ACM SIGKDD\par Conference.\par \par The Stanford Linear Accelerator C http://www.downhi.com/txt/PMhieofiFTRK.html enter SPIRES-HEP database has been\par comprehensively cataloguing the High Energy Particle Physics (HEP) literature\par online since 1974, and indexes more than 500,000 high-energy physics related\par articles including their full citation tree.\par \par The network contains a citation graph of the hep-th portion of the arXiv. The\par units names are the arXiv IDs of papers; the relation is X cites Y . Note that\par revised papers may have updated citations. As such, citations may refer to\par future papers, i.e. a paper may cite another paper that was published after the\par first paper.\par \par The SLAC/SPIRES dates for all hep-th papers are given. Some older papers were\par uploaded years after their intial publication and the arXiv submission date\par from the abstracts may not correspond to the publication date. An alternative\par date has been provided from SLAC/SPIRES that may be a better estimate for the\par initial publication of these old papers.\par \par The first version of data was updated on May 12, 2003.\par \par hep-th.net X cites Y relation, first version. hep-th-new.net X cites Y\par relation, updated version. date-new.vec SLAC date of paper was transformed to\par the number of days since August 1, 1991, updated version. year-new.vec year\par from the SLAC date of paper, updated version.\par \par References\par \par 1. KDD Cup 2003\par \par 2. arXiv\par \par Transformed in Pajek format by V. Batagelj, 26. July 2003\par\par \par \par ================================================================================\par ==> IMDB/IMDB.txt <==\par ================================================================================\par \par Pajek data set: IMDB, the Internet Movie Database, http://www.imdb.com.\par \par -------------------------------------------------------------------------------'\par Pajek network converted to sparse adjacency matrix for inc http://www.downhi.com/txt/PMhieofiFTRK.html lusion in UF sparse\par matrix collection, Tim Davis. For Pajek datasets, See V. Batagelj & A. Mrvar,\par http://vlado.fmf.uni-lj.si/pub/networks/data/.\par -------------------------------------------------------------------------------\par A(i,j)=1 if actor j played in movie i. colname(j,:) is the name of the actor.\par Column j = 362,181 is Kevin Bacon. Year of movie i is year(i).\par category(i) gives the category of movie i, use code(category(i),:).\par Note that movie names are not provided.\par \par 1: Drama\par 2: Short\par 3: Documentary\par 4: Comedy\par 5: Western\par 6: Family\par 7: Mystery\par 8: Thriller\par 9: -\par 10: Music\par 11: Crime\par 12: Sci-Fi\par 13: Horror\par 14: War\par 15: Fantasy\par 16: Romance\par 17: Adventure\par 18: Animation\par 19: Action\par 20: Musical\par 21: Film-Noir\par 99: Unknown.\par \par Remember that in MATLAB, A(i,:) is slow to compute; A(:,i) is faster. If you\par want row i of a sparse matrix, access the ith column of the transpose instead.\par -------------------------------------------------------------------------------\par \par A "Bacon number" is a measure of separation in the graph. Kevin Bacon has\par a Bacon number of zero. Any actor who played in a movie with Kevin Bacon\par has a Bacon number of 1. In general, an actor has a Bacon number equal to\par 1 the minimum Bacon number of any actor he or she has been in a movie with.\par \par A similar definition can be extended to movies; a movie in which Kevin Bacon\par appears has a Bacon number of 0. In general, the Movie-Bacon number is\par the smallest Bacon number of any actor in that movie.\par \par The following code was used to compute the Bacon numbers:\par \par Bacon = Problem.aux.KevinBacon ;\par A = Problem.A ;\par [m n] = size (A) ;\par C = [speye(m) A ; A' speye(n)] ;\par x = zeros (m n,1) ;\par B = inf * ones (m n,1 http://www.downhi.com/txt/PMhieofiFTRK.html ) ;\par x (m Bacon) = 1 ;\par B (m Bacon) = 0 ;\par tlen = 1 ;\par for k = 1:m n\par x = x C*x ;\par t = find (x) ;\par if (tlen == length (t))\par break\par end\par tlen = length (t) ;\par B (t) = min (B (t), k) ;\par end\par MovieBacon = (B (1:m) - 1) / 2 ;\par ActorBacon = B (m 1:end) / 2 ;\par \par Note that the movie names have been intentionally omitted from this version of\par the data in the UF Sparse Matrix Collection, as has the name of movie code 9.\par \par The above MATLAB code is not terribly efficient; it makes k passes over the\par matrix, each taking O(nnz(A)) time. Fortunately, k is a small constant (8).\par A proper breadth-first search would take O(nnz(A)), regardless of k.\par \par ================================================================================\par ==> Journals/Journals.txt <==\par ================================================================================\par \par Pajek datasets Slovenian magazines and journals 1999 and 2000\par \par Dataset Journals\par \par Description\par \par Revije.net - valued network with 124 vertices Revije.clu - partition with 124\par vertices Revije.paj - Pajek project file with complete dataset.\par\par \par \par complete dataset (ZIP, 3K)\par \par Background\par \par Over 100.000 people have been asked which magazines and journals they read\par (survey conducted in 1999 and 2000, source CATI Center Ljubljana). They listed\par 124 different magazines and journals. The collected data can be represented as\par 2-mode network:\par \par Delo Dnevnik Sl.novice ...\par Reader1 X X ...\par Reader2 X ...\par Reader3 X ...\par ............ ..... ....... ......... ...\par \par Obtaining 1-mode network\par \par From 2-mode network http://www.downhi.com/txt/PMhieofiFTRK.html reader/journal we generated ordinary network, where the\par vertices are journals\par \par * undirected edge with value a between journals means the number of readers\par * of both journals. loop on selected journal means the number of all\par * readers that read this journal.\par \par Obtained matrix (A):\par \par Delo Dnevnik Sl.novice ...\par Delo 20714 3219 4214 ...\par Dnevnik 15992 3642 ...\par Sl. novice 31997 ...\par ........ ...... ..... ..... ...\par \par The second ordinary network on readers would be huge (more than 100.000\par vertices) containing large cliques (readers of particular journal).\par \par History\par \par 1. Transformed in journal X journal matrix, 26. December 2000.\par \par 2. Transformed in Pajek format by V. Batagelj, 21. December 2003.\par \par References\par \par 1. Dagstuhl seminar: Link Analysis and Visualization, Dagstuhl 1-6. July 2001\par \par Pajek Data; 21. December 2003\par\par \par ================================================================================\par ==> KEDS/KEDS.txt <==\par ================================================================================\par \par NOTE: the KEDS data has not been included.\par \par Pajek datasets KEDS The Kansas Event Data System\par \par Dataset KEDS\par \par Description\par \par GulfLDays.net directed multirelational temporal network with 174 vertices and\par 57131 arcs. From 'leads' Gulf event data, granularity is 1 day.\par \par GulfLMonths.net directed multirelational temporal network with 174 vertices and\par 57131 arcs. From 'leads' Gulf event data, granularity is 1 month.\par \par GulfLDow.net directed multirelational temporal network with 174 vertices and\par 57131 arcs. From 'leads' Gulf event data, in day of the week classes.\par \p http://www.downhi.com/txt/PMhieofiFTRK.html ar GulfADays.net directed multirelational temporal network with 202 vertices and\par 304401 arcs. From Gulf event data, granularity is 1 day.\par \par GulfAMonths.net directed multirelational temporal network with 202 vertices and\par 304401 arcs. From Gulf event data, granularity is 1 month.\par \par LevantDays.net directed multirelational temporal network with 485 vertices and\par 196364 arcs. From Levant event data, granularity is 1 day.\par \par LevantMonths.net directed multirelational temporal network with 485 vertices\par and 196364 arcs. From Levant event data, granularity is 1 month.\par \par BalkanDays.net directed multirelational temporal network with 325 vertices and\par 78667 arcs. From Balkan event data, granularity is 1 day.\par \par BalkanMonths.net directed multirelational temporal network with 325 vertices\par and 78667 arcs. From Balkan event data, granularity is 1 month.\par\par \par \par GulfLDays.net (ZIP, 239K)\par \par GulfLMonths.net (ZIP, 197K)\par \par GulfLDow.net (ZIP, 213K)\par \par GulfADays.net (ZIP, 1078K)\par \par GulfAMonths.net (ZIP, 941K)\par \par LevantDays.net (ZIP, 855K)\par \par LevantMonths.net (ZIP, 735K)\par \par BalkanDays.net (ZIP, 335K)\par \par BalkanMonths.net (ZIP, 288K)\par \par Background\par \par KEDS - The Kansas Event Data System uses automated coding of English-language\par news reports to generate political event data focusing on the Middle East,\par Balkans, and West Africa. These data are used in statistical early warning\par models to predict political change. The ten-year project is based in the\par Department of Political Science at the University of Kansas; it has been funded\par primarily by the U.S. National Science Foundation. KEDS data sets from KEDS\par data collection.\par \par Gulf data set: This data set covers the states of the Gulf region and the\par Arabian peninsula for the period 15 April 1979 to http://www.downhi.com/txt/PMhieofiFTRK.html 31 March 1999. The source\par texts prior to 10 June 97 were located using a NEXIS search command\par specifically designed to return relevant data. There are two versions of the\par data: a set coded from the lead sentences only (57,000 events), and a set coded\par from full stories (304,000 events). There are some errors in the GulfAll data.\par Events 118196 and 118197 have REUT-0 in place of the date; and in event 173526\par the first actor is missing. In Gulf99All.dat the wrong dates are replaced with\par 890319, and the incomplete event is skiped.\par \par Levant data set: Folder containing WEIS-coded events (N=196,337) for dyadic\par interactions within the following set of countries: Egypt, Israel, Jordan,\par Lebanon, Palestinians, Syria, USA, and USSR/Russia. Coverage is April 1979 to\par June 2004. TABARI coding dictionaries are also included. There are some errors\par (333) in data set - relation codes 012], O24, O53, 213] are replaced with 012,\par 024, 053, 213 in Levant.dat. Some events don't have description codes - they\par are marked with *** in relation labels in *.net files.\par \par Balkans data set, 1989-2003: Folder containing WEIS-coded events (N = 78,667)\par for the major actors (including ethnic groups) involved in the conflicts in the\par former Yugoslavia. Coverage is April 1989 through July 2003. TABARI coding\par dictionaries are included in the folder. There are some errors (197) in data\par set - relation codes ---], O24, O53 are replaced with 000, 024, 053 in\par Balkan.dat. Some events don't have description codes - they are marked with ***\par in relation labels in *.net files.\par \par The original data sets are on MAC files. They should be saved as PC files\par before processing. \par \par History\par \par 1. Program Recode (in Delphi) by Vladimir Batagelj, Ljubljana, July 25, 2003\par \par 2. Program KEDSrec adapted for KEDS from Recode; Gul http://www.downhi.com/txt/PMhieofiFTRK.html f (leads) data recoded\par into Pajek's format, by Vladimir Batagelj, Ljubljana, November 3, 2003\par \par 3. Support for multiple relations networks added to program KEDSrec by\par Vladimir Batagelj, Ljubljana, November 22, 2004\par \par 4. KedsR - functionality of KEDSrec implemented in R; Gulf (leads) data\par recoded into Pajek's format (days, months, day of the week), by Vladimir\par Batagelj, Ljubljana, November 27, 2004\par \par 5. WeisR - commands from KedsR adapted for WEIS format (similar to KEDS but\par TAB delimited); Balkan and Levant data recoded into Pajek's format (days,\par months), by Vladimir Batagelj, Ljubljana, November 28, 2004\par \par 6. Gulf99All is a large data set - sapply commands in KedsR had to be\par replaced by while loops; Gulf (all) data recoded into Pajek's format (days,\par months), by Vladimir Batagelj, Ljubljana, November 29, 2004\par\par \par References\par \par 1. StuffIt - uncompress program for SIT files\par \par Pajek Data; 29. November 2004 / 24. November 2004\par\par \par ================================================================================\par ==> Mixed/Mixed.txt <==\par ================================================================================\par\par \par Pajek datasets from different sources\par \par US power grid\par \par Dataset: USpowerGrid\par \par US power grid - unweighted network from Panayiotis Tsaparas' page. Adapted for\par Pajek by V. Batagelj, March 19, 2006\par \par File: USpowerGrid.net - undirected network with 4941 vertices and 6594 edges\par\par \par \par Kleinberg's web graphs\par \par Dataset: California - Pages matching the query "California".\par \par This graph was constructed by expanding a 200-page response set to a search\par engine query 'California', as in the hub/authority algorithm. Obtained from Jon\par Kleinberg's page. Adapted for Pajek by V. Batagelj, March 19, http://www.downhi.com/txt/PMhieofiFTRK.html 2006\par \par File: California.net - directed network with 9664 vertices and 16150 arcs.\par\par \par References: Amy N. Langville and Carl D. Meyer: A Reordering for the PageRank\par problem (March 2004) PDF J. Kleinberg. Authoritative sources in a hyperlinked\par environment. Proc. 9th ACM-SIAM Symposium on Discrete Algorithms, 1998.\par Extended version in Journal of the ACM 46(1999). PDF\par \par Dataset: Epa - Pages linking to www.epa.gov.\par \par This graph was constructed by expanding a 200-page response set to a search\par engine query, as in the hub/authority algorithm. Obtained from Jon Kleinberg's\par page. Adapted for Pajek by V. Batagelj, March 19, 2006\par \par File: Epa.net - directed network with 4772 vertices and 8965 arcs.\par\par \par Stanford web graphs\par \par Dataset: StanfordWeb - Stanford Web Matrix.\par \par This graph was obtained from a September 2002 crawl (281903 pages, 2382912\par links). The matrix rows represent the inlinks of a page, and the columns\par represent the outlinks. Downloaded from Sepandar D. Kamvar's page. Adapted for\par Pajek by V. Batagelj, March 19, 2006\par \par File: StanfordWeb.net - directed network with 281903 vertices and 2382912 arcs.\par\par \par References: Sepandar D. Kamvar, Taher H. Haveliwala, Christopher D. Manning,\par and Gene H. Golub, "Exploiting the Block Structure of the Web for Computing\par PageRank." Preprint (March, 2003). PDF\par \par Dataset: StanfordBerkeleyWeb - Stanford-Berkeley Web Matrix.\par \par This graph was obtained from a December 2002 crawl (685230 pages, 8006115\par links). The matrix rows represent the inlinks of a page, and the columns\par represent the outlinks. Downloaded from Sepandar D. Kamvar's page. Adapted for\par Pajek by V. Batagelj, March 19, 2006\par \par NOTE by Tim Davis: nodes 683,447 to 685,230 in the Stanford Berkeley web data,\par discussed above, are not part of http://www.downhi.com/txt/PMhieofiFTRK.html the true results. Kamvar's MATLAB script for\par processing the data deletes those nodes. This graph is already in the UF\par Sparse Matrix Collection, in the correct size. Also note that Kamvar doesn't\par consider multiple links from page i to j to be significant. Thus, duplicate\par edges (i,j) in this graph should be ignored. The graph in the UF collection is\par thus binary.\par \par File: StanfordBerkeleyWeb.net - directed network with 685230 vertices and\par 8006115 arcs.\par\par \par References: Sepandar D. Kamvar, Taher H. Haveliwala, Christopher D. Manning,\par and Gene H. Golub, "Exploiting the Block Structure of the Web for Computing\par PageRank." Preprint (March, 2003). PDF\par \par World City Network\par \par Dataset: WorldCities\par \par These data constitute the empirical basis for measuring the world city network\par as described in P.J. Taylor (2004) World City Network: A Global Urban Analysis\par (London: Routledge). The data were produced by P.J. Taylor and G. Catalano.\par These data were collected in 2000. They are the service values (indicating the\par importance of a city in the office network of a firm) of 100 global service\par firms distributed across 315 cities worldwide. All firms supply advanced\par producer services (accountancy, advertising, banking/finance, insurance, law,\par and management consultancy) through offices in at least 15 cities (including at\par least one in each of Pacific Asia, western Europe and northern America). The\par following coding information is given to help understand and evaluate the data.\par Downloaded from GaWC Data Set 11 page. Transformed in Pajek format by V.\par Batagelj, March 20, 2006\par \par Files:\par \par WorldCities.net - two-mode network with 415=315 100 vertices and 7518 arcs.\par \par WorldCities.mat - same network in Pajek matrix format.\par \par WorldCities.net - type of service partition: 0 - c http://www.downhi.com/txt/PMhieofiFTRK.html ity; 1 - accountancy; 2 -\par advertising; 3 - banking/finance; 4 - insurance; 5 - law; 6 - management\par consultancy.\par\par \par \par References: Taylor, P.J. (2004): World city network: a global urban analysis.\par London and New York: Routledge.\par \par Pajek Data; 19. March 2006\par\par \par ================================================================================\par ==> ND/ND.txt <==\par ================================================================================\par\par \par Pajek datasets Notre Dame Self-Organized Networks Database\par \par Datasets NDwww, NDactors, NDyeast\par \par Description\par \par Notre Dame Self-Organized Networks:\par \par 1. NDwww.net directed network with 325729 vertices and 1497135 arcs (27455\par loops); page X is linked to page Y.\par \par 2. NDactors.net undirected two-mode network with 520223 vertices (392400\par players, 127823 movies) and 1470418 edges; player X plays in movie Y.\par \par 3. NDyeast.net undirected network with 2114 vertices and 2277 edges (74\par loops); protein X interacts with protein Y.\par\par \par \par NDwww.net (ZIP, 2050K) NDactors.net (ZIP, 4150K) NDyeast.net (ZIP, 7K)\par \par Background\par \par The networks ND*.net are based on the files from Notre Dame Self-Organized\par Networks Database. To transform the data into Pajek format: vertex 0 was\par replaced by the vertex number equal to the number of vertices in a network;\par Pajek keywords were inserted; and the network was saved in the short (as lists\par of neighbors) format.\par \par 1. World-Wide-Web:: Each number represents webpage within nd.edu domain.\par Arcs: From page -> To page R�ka Albert, Hawoong Jeong and Albert-L�szl�\par Barab�si: Diameter of the World Wide Web, Nature 401, 130 (1999) [ PDF ] See\par also a decompostion of this network in V. Batgelj, A. Mrvar: How to analyze\par large networks with Pajek?\par \par http://www.downhi.com/txt/PMhieofiFTRK.html 2. Actor: Actor network data: (based on www.imdb.com) In the original ND\par network file: each line corresponds to one movie, each number represents actor:\par number_1 number_2 ... number_k (k actors who play in the same movie).\par Albert-L�szl� Barab�si, R�ka Albert: Emergence of scaling in random networks,\par Science 286, 509 (1999) [ PDF ]\par \par 3. Protein Interaction Network for Yeast: Each number represents protein in\par protein interaction network of yeast. Edges: From protein -> To protein. For\par other datasets used in supplementary material, please refer indicated\par references. Hawoong Jeong, Sean Mason, Albert-L�szl� Barab�si and Zolt�n N.\par Oltvai: Centrality and lethality of protein networks, Nature 411, 41 (2001) [\par PDF ] See also Yeast data\par \par History\par \par 1. Notre Dame Networks Database put on WWW by the Notre Dame team, 2001;\par \par 2. 23-25. July 2001: ND nets transformed in Pajek format by V. Batagelj.\par \par 3. 23. May 2004: ND nets in Pajek format transformed in short (lists of\par neighbors) Pajek format by V. Batagelj.\par \par References\par \par 1. Self-Organized Networks Database, University of Notre Dame.\par \par Copyright Extract from the Notre Dame Networks Database page: "... Feel free to\par use these data in your research." Mail to Hawoong Jeong (author of original ND\par networks).\par \par Pajek Data; 23. May 2004\par\par \par ================================================================================\par ==> ODLIS/ODLIS.txt <==\par ================================================================================\par\par \par Pajek datasets\par \par ODLIS Online Dictionary of Library and Information Science\par \par Dataset odlis\par \par Description\par \par odlis.net directed network with 2909 vertices and 18419 arcs (5 loops).\par\par \par \par odlis.net (ZIP, 62K) Background\par \pa http://www.downhi.com/txt/PMhieofiFTRK.html r The network Odlis.net is based on the ODLIS: Online Dictionary of Library and\par Information Science. version December 2000.\par \par ODLIS is designed to be a hypertext reference resource for library and\par information science professionals, university students and faculty, and users\par of all types of libraries. The primary criterion for including a new term is\par whether a librarian or other information professional might reasonably be\par expected to encounter it at some point in his (or her) career, or be required\par to know its meaning in the course of executing his or her responsibilities as a\par librarian. The vocabulary of publishing, printing, book history, literature,\par and computer science has been included when, in the author's judgment, a\par definition might prove helpful, not only to library and information\par professionals, but also to laypersons.\par \par An arc (X,Y) from term X to term Y exists in the network iff in the ODLIS\par dictionary the term Y is used to describe the meaning of term X.\par \par ODLIS is the work of Joan M. Reitz, Assistant Professor/Instruction Librarian\par at the Ruth A. Haas Library, Western Connecticut State University (WCSU) in\par Danbury, CT. Ms. Reitz holds an M.L.I.S. degree (1991) from the University of\par Washington in Seattle and an M.A. degree (1998) in European History from\par Western Connecticut State University. Her primary research interests are the\par history of the book and history of political and social revolutions. \par \par History\par \par 1. ODLIS began at the Haas Library in 1994 as a five-page photocopied\par handout intended for undergraduates not fluent in English, and students with\par limited exposure to library terminology.\par \par 2. In 1996, it was expanded and converted to HTML format for installation on\par the WCSU Libraries HomePage under the title Hypertext Library Lingo: A Glossary\pa http://www.downhi.com/txt/PMhieofiFTRK.html r of Library Terminology.\par \par 3. In 1997, many more hypertext links were added and the format improved in\par response to suggestions from users.\par \par 4. During the summer of 1999, several hundred terms and definitions were\par added, and a generic version created which omitted all references to the\par specific conditions and practices at the Haas Library.\par \par 5. In the fall of 1999, the glossary was expanded to 1,800 terms, renamed to\par reflect its extended scope, and copyrighted.\par \par 6. In February, 2000, ODLIS was indexed in Yahoo! under "Reference -\par Dictionaries - Subject." It is also indexed in the WorldCat database in OCLC\par FirstSearch.\par \par 7. During the year 2000, the dictionary was expanded to 2,600 terms. On\par average, it has received over 6,200 visits per month since January 2, 2000.\par \par 8. December 2000: ODLIS transformed in Pajek format and 'cleaned' by A.\par Mrvar and V. Batagelj.\par \par References\par \par 1. Joan M. Reitz (2002): ODLIS: Online Dictionary of Library and Information\par Science.\par \par Pajek Data; 27. January 2004\par\par \par ================================================================================\par ==> Patents/Patents.txt <==\par ================================================================================\par\par \par Pajek datasets Patents The NBER U.S. Patent Citations Data File\par \par Dataset patents\par \par Description\par \par patents.net directed network with 3,774,768 vertices and 16,522,438 arcs (1\par loop).\par\par \par \par NET, NAM, Year, Date, Cat, Class, Country, Subcat, AppYear Patents NET / main\par subnetwork Background\par \par The network Patents is based on the The NBER U.S. Patent Citations Data File,\par version 2001.\par \par These data comprise detail information on almost 3 million U.S. patents granted\par between January 1963 and Dec http://www.downhi.com/txt/PMhieofiFTRK.html ember 1999, all citations made to these patents\par between 1975 and 1999 (over 16 million), and a reasonably broad match of\par patents to Compustat (the data set of all firms traded in the U.S. stock\par market).\par \par These data are described in detail in Hall, B. H., A. B. Jaffe, and M.\par Tratjenberg (2001). "The NBER Patent Citation Data File: Lessons, Insights and\par Methodological Tools." NBER Working Paper 8498. ALL USERS OF THESE DATA SHOULD\par READ THIS PAPER, AND SHOULD CITE IT AS THE SOURCE OF THE DATA\par \par Further documentation on uses of the patent citation data, including the\par methodology paper and a CD containing the complete dataset itself, is available\par in the book Patents, Citations and Innovations: A Window on the Knowledge\par Economy by Adam Jaffe and Manuel Trajtenberg, MIT Press, Cambridge (2002). The\par book may be ordered from MIT Press. ISBN 0-262-10095-9.\par \par History\par \par 1. Data produced by United States Patent and Trademark Office\par \par 2. Hall, B. H., A. B. Jaffe, and M. Tratjenberg prepared the NBER dataset,\par 2001\par \par 3. July 10, 2003: NBER data transformed in Pajek format by V. Batagelj.\par\par \par References\par \par 1. Hall, B. H., A. B. Jaffe, and M. Tratjenberg (2001): The NBER U.S. Patent\par Citations Data File.\par \par 2. United States Patent and Trademark Office, Patent Number Search\par \par 3. Batagelj V. (2003): Efficient Algorithms for Citation Network Analysis.\par \par Pajek Data; 27. January 2004\par \par -------------------------------------------------------------------------------\par NOTE regarding conversion for UF sparse matrix collection: in original the data\par there are 14,973,817 edges (unweighted). Of this, 3050 are duplicates \par This graph is binary; the duplicates have been removed. \par Also, the original data has auxilia http://www.downhi.com/txt/PMhieofiFTRK.html ry data for all 6,009,554 US Patents in this\par time period. This patent network has only 3,774,768 patents, and the auxiliary\par data (appyear, class, etc.) is matched here to the nodes of the graph. \par \par ================================================================================\par ==> Roget/Roget.txt <==\par ================================================================================\par\par \par Pajek datasets Roget Roget's Thesaurus, 1879\par \par Dataset Roget\par \par Description\par \par roget.net directed network with 1022 vertices and 5075 arcs (1 loop); word X is\par related to word Y.\par\par \par \par Roget.net (ZIP, 17K)\par \par Background\par \par The network Roget.net is based on the file roget.dat from the Stanford\par GraphBase that contains cross-references in Roget's Thesaurus, 1879.\par \par Dr. Peter Mark Roget (1779-1869) philologist, scientist, physician. The name\par Roget could soon become a virtual synonym for the word "synonym". For those who\par use Roget's Thesaurus it is one of the three most important books ever\par printed...along with The Bible and Webster's Dictionary. In order to\par communicate one's exact intention...or one's precise meaning, the Thesaurus,\par being a list of synonyms or verbal equivalents, is a necessary tool. The first\par draft of the Thesaurus was written in 1805, two years before Webster started on\par his dictionary. However for a period of 47 years Dr. Roget used his manuscript\par as his personal, secret, treasure trove. Not until he was 73 years old did he\par decide to reveal and publish this great manuscript.\par \par Since 1852, Roget's Thesaurus has never been out of print. In fact, each\par succeeding edition has increased the popularity of the work. The original\par 15,000 words included in the 1805 manuscript has increased to over a quarter of\par a million in the 1992 edition (the te http://www.downhi.com/txt/PMhieofiFTRK.html nth printing). With such an increase in\par size, it is encouraging to notice that the basic content still remains\par intact..... for example, where the 1805 Thesaurus traces the word: existence:\par "Ens, entity, being, existence, essence...", the 1992 Thesaurus contains\par existence: "existence, being, entity, ens,...essence..."\par \par Each vertex of the graph corresponds to one of the 1022 categories in the 1879\par edition of Peter Mark Roget's Thesaurus of English Words and Phrases, edited by\par John Lewis Roget. An arc goes from one category to another if Roget gave a\par reference to the latter among the words and phrases of the former, or if the\par two categories were directly related to each other by their positions in\par Roget's book. For example, the vertex for category 312 (`ascent') has arcs to\par the vertices for categories 224 (`obliquity'), 313 (`descent'), and 316\par (`leap'), because Roget gave explicit cross-references from 312 to 224 and 316,\par and because category 312 was implicitly paired with 313 in his scheme.\par \par History\par \par 1. Original Roget's Thesaurus was published in 1852.\par \par 2. Peter's son John Luis Roget published the second, improved edition in\par 1879.\par \par 3. Project Gutenberg Roget's Thesaurus (1911 edition) put into electronic\par format in 1991.\par \par 4. Graph Roget.dat of cross-references based on the second edition was\par produced for Stanford GraphBase (SGB) in 1992/3.\par \par 5. MICRA (Pat Cassidy) prepared the electronic version of the 1911 Roget's\par Thesaurus that is widely available on the internet.\par \par 6. SGB Roget.dat transformed in Pajek format: A. Mrvar, 5. December 1996.\par \par References\par \par 1. Peter Mark Roget: Roget's Thesaurus of English Words and Phrases\par \par 2. Project Gutenberg: Roget's Thesaurus\par \par 3. Donald E. Knuth: The Stanford GraphBa http://www.downhi.com/txt/PMhieofiFTRK.html se: A Platform for Combinatorial\par Computing . New York: ACM Press, 1993\par \par 4. The Stanford GraphBase: roget.dat, version 15.6.1993\par \par 5. Pat Cassidy: MICRA / Factotum\par \par 6. CIDE (Collaborative International Dictionary of English), GNU 1996-2002\par \par Pajek Data; 23. January 2004\par\par \par ================================================================================\par ==> Sandi/Sandi.txt <==\par ================================================================================\par \par REFERENCES\par \par DATASET SANDI\par \par DESCRIPTION 2-mode 674�314 network.\par \par BACKGROUND These data were obtained from the bibliography of the book Imrich W,\par Klavžar S. (1999) Graph products. The result is a author-by-paper 2-mode\par network: arc (i,j) - author i is the (co)author of the paper j.\par \par DERIVED DATA AUTHORS authors network\par \par REFERENCES\par \par * Imrich W, Klavžar S. (1999). Graph products. References.\par\par \par \par ================================================================================\par ==> Stranke94/Slovene_Parties.txt <==\par ================================================================================\par \par Pajek datasets Slovene Parliamentary Parties 1994\par \par Dataset Stranke94\par \par Description\par \par Stranke94.net - valued signed network with 10 vertices\par\par \par \par Stranke94.net Background\par \par Relations between Slovene parliamentary political parties:\par \par * SKD - Slovene Christian Democrats;\par \par * ZLSD - Associated List of Social Democrats;\par \par * SDSS - Social Democratic Party of Slovenia;\par \par * LDS - Liberal Democratic Party;\par \par * ZSESS - first of two Green Parties, separated after 1992 elections;\par \par * ZS - second Green Party;\par \par * DS - Democratic Party;\par \par * SLS - Slovene http://www.downhi.com/txt/PMhieofiFTRK.html People's Party;\par \par * SNS - Slovene National Party;\par \par * SPS SNS - a group of deputies, former members of SNS, separated after\par * 1992 elections\par \par were estimated by the members of the Slovene National Parliament. So the\par respondents were well informed and competent to give such estimations. In the\par questionnaire designed by a group of experts on Parliament activities, some\par questions about the political space and its dimensions were included and the\par following question about relations between parliamentary political parties:\par \par If various criteria (or various dimensions of the political space) are\par taken into account, some parties are by average closer than others. How\par would you personally estimate distances between pairs of parties in the\par political space?\par \par Please, estimate the distance between each pair of parties on the scale\par from -3 to 3, where:\par \par -3 means that parties are very dissimilar;\par \par -2 means that parties are quite dissimilar;\par \par -1 means that parties are dissimilar;\par \par 0 means that parties are neither dissimilar nor similar (somewhere in\par between);\par \par 1 means that parties are similar;\par \par 2 means that parties are quite similar;\par \par 3 means that parties are very similar.\par \par To collect estimations each respondent was given a 10-party by 10-party table\par with empty cells in the upper triangle. The diagonal and the lower triangle\par were coloured in black. Each respondent had to estimate relations between 45\par pairs of parties.\par \par The measures of central tendency were computed on the basis of the estimations\par given by 72 out of 90 members of the Parliament. 17 members of the Parliament\par were not available at the time of interviewing and one refused respond. 64\par respo http://www.downhi.com/txt/PMhieofiFTRK.html ndents out of the 72 estimated all 45 requested party relations. Only two\par out of 8 respondents with missing values have a large number of missing values\par (namely 40), the rest of them have 5 to 10 missing values. As far as the\par parties (variables) are considered, there are from 0 to 8 missing values and a\par recognisable pattern: relations involving ZS (which had no representatives in\par the Parliament at the time of interviewing) include from 6 to 8 missing values,\par others from 0 to 3 missing values.\par \par The weights of arcs in the network are averages of values multiplied by 100 and\par rounded. The missing values were excluded. \par \par History\par \par 1. Stran format by V. Batagelj, 19. October 1994.\par \par 2. Transformed in Pajek format by V. Batagelj, 15. February 2004.\par \par References\par \par 1. Samo Kropivnik and Andrej Mrvar: An Analysis of the Slovene Parliamentary\par Parties Network. Developments in Statistics and Methodology. (A. Ferligoj,\par A. Kramberger, editors) Metodološki zvezki 12, FDV, Ljubljana, 1996, p.\par 209-216.\par \par 2. Patrick Doreian and Andrej Mrvar(1996): A Partitioning Approach to\par Structural Balance. Social Networks, 18, p. 149-168.\par \par Pajek Data; 15. February 2004\par\par \par ================================================================================\par ==> Tina/Tina.txt <==\par ================================================================================\par \par Pajek datasets\par Student Government of the University of Ljubljana / 1992\par \par Dataset Tina\par \par Description\par \par DisCal.net - network with 11 vertices and 41 arcs\par DisCog.net - network with 11 vertices and 48 arcs\par AskCal.net - network with 11 vertices and 29 arcs\par AskCog.net - network with 11 vertices and 36 arcs\par AnsCalT.net - network with 11 vertices and 41 arcs\par AnsCogT.net - http://www.downhi.com/txt/PMhieofiFTRK.html network with 11 vertices and 42 arcs\par DisCalSn.net - network with 11 vertices and 41 arcs\par Tina.paj - Pajek project file with complete dataset.\par\par \par \par complete dataset (ZIP, 3K)\par DisCalSn.net\par Background\par In the experiment two alternative methods were used for collection of network data:\par \par Recall: Members of the group were asked to identify the members of their\par egocentric networks by memory. The criteria for enumeration was frequency\par of the recalled communications.\par \par Recognition: The list of all members of the group was given to each\par member. They were first asked to identify who they communicate with and\par than to select the persons they communicate with most often.\par \par The number of listed persons was not limited in any method.\par \par The analyzed network consisted of communication interactions among twelve\par members and advisors of the Student Government at the University in Ljubljana.\par The results of the measurement are not real interactions among actors but\par cognition about communication interactions. Data were collected with face to\par face interviews which lasted from 20 to 30 minutes. Interviews were conducted\par in May 1992.\par \par Communication flow among actors was identified by three questions:\par \par Who of the members and advisors of the Student government do you (most\par often) informally discuss with?\par \par Which members and advisors of the Student Government do you (most often)\par ask for an opinion?\par \par Which of the members and advisors of the Student Government (most often)\par ask you for an opinion?\par \par The content of the communication flow was limited to the matters of the Student\par Government. The time frame was also defined: questions were referred to the six\par months period (from the formation of the government http://www.downhi.com/txt/PMhieofiFTRK.html to the day of the\par interview).\par \par Only one respondent listed all the others under the recognition method for two\par relations (discussion, asking for an opinion). For that respondent the first\par group was defined arbitrary. The cut point was determined by the average number\par of selected persons at the recognition method (3.5 for the first relation and\par 4.5 for the second).\par \par One respondent refused to cooperate in the experiment. As he was not considered\par in the analysis, the network consists of eleven actors.\par\par \par History\par\par \par 1. Transformed in Stran format by V. Batagelj, 28. July 1993.\par 2. Transformed in Pajek format by V. Batagelj, 8. August 2003.\par \par Files\par \par DisCal.net - discussion, recall\par DisCog.net - discussion, recognition\par AskCal.net - asking for an opinion, recall\par AskCog.net - asking for an opinion, recognition\par AnsCalT.net - being asked for an opinion, recall (transposed)\par AnsCogT.net - being asked for an opinion, recognition (transposed)\par DisCalSn.net - discussion, recall / short names\par Tina.paj - Pajek project file with complete dataset.\par References\par \par 1. Valentina Hlebec: Recall Versus Recognition: Comparison of the Two\par Alternative Procedures for Collecting Social Network Data. Developments in\par Statistics and Methodology. (A. Ferligoj, A. Kramberger, editors)\par Metodološki zvezki 9, FDV, Ljubljana, 1993, p. 121-129.\par \par Pajek Data; 8. August 2003\par\par \par ================================================================================\par ==> WassermanFaust/WassermanFaust.txt <==\par ================================================================================\par \par NOTE: the Wasserman/Faust networks have not been included.\par\par \par Pajek datasets\par from the book\par Social Network Analysis: Methods and Applications\par Wasserman http://www.downhi.com/txt/PMhieofiFTRK.html and Faust, 1994\par Wasserman and Faust datasets\par \par Dataset WaFa\par \par There are five network examples: Krackhardt's high-tech managers, Padgett's\par Florentine families, Freeman's EIES network, Countries trade network, and\par Galaskiewicz's CEOs and clubs network.\par \par Description\par \par HighTech.paj - multirelational network with 21 vertices and 190 102 20 arcs\par Padgett.paj - multirelational network with 16 vertices and 30 40 arcs\par EIES.paj - multirelational network with 48/32 vertices and 695 830/460 arcs\par Trade.paj - multirelational network with 24 vertices and 307 307 310 135 369\par arcs\par CEOs.net - two-mode network with 41 vertices and 98 edges.\par\par \par Background\par \par Complete descriptions of these data, including references for the original\par sources of the data, can be found in Chapter 2 (pages 59- 66) and Appendix B\par (pages 738-755) of Wasserman and Faust.\par \par The original data in ASCII, UCINET and KrackPlot formats are available at\par INSNA. \par \par History\par \par 1. INSNA page.\par 2. Transformed in Pajek format by V. Batagelj, 15. January 2005.\par \par Files\par \par HighTech.paj - Krackhardt's High-tech managers\par Padgett.paj - Padgett's Florentine Families\par EIES.paj - Freeman's EIES network\par Trade.paj - Countries trade data\par CEOs.net - Galaskiewicz's CEOs and clubs.\par References\par \par 1. Stanley Wasserman, Katherine Faust: Social Network Analysis: Methods and\par Applications. CUP, 1994.\par \par Pajek Data; 15. January 2005\par\par \par ================================================================================\par ==> Wordnet/Wordnet.txt <==\par ================================================================================\par\par \par NOTE: this is a binary graph in the Pajek dataset, but where each edge has a\par label (not a weight) in the range 1 to 9. The http://www.downhi.com/txt/PMhieofiFTRK.html following labels are used:\par \par 1 hypernym pointer\par 2 entailment pointer\par 3 similar pointer\par 4 member meronym pointer\par 5 substance meronym pointer\par 6 part meronym pointer\par 7 cause pointer\par 8 grouped pointer\par 9 attribute pointer\par \par This is not a multigraph. There are no edges (i,j) between the same nodes\par with the same label. Thus, in the sparse matrix, the edge weight A(i,j)\par represents the label 1 through 9 of edge (i,j). No loss of information\par occurs in this translation. The above table is in aux.edgecode(1:9,:).\par Each node is a word in a dictionary. aux.category(i) gives the category\par of the word:\par 1: n (noun?) 63099 words\par 2: v (verb?) 4496 words\par 3: a (adjective?) 5501 words\par 4: r (?) 2846 words\par 5: s (?) 6728 words.\par\par \par ================================================================================\par ==> yeast/yeast.txt <==\par ================================================================================\par \par Pajek datasets\par Protein-protein interaction network in budding yeast\par \par Dataset Yeast\par \par Description\par \par yeastS.net network with 2361 vertices and 7182 edges (536 loops).\par yeastL.net network with 2361 vertices and 7182 edges (536 loops).\par yeast.clu partition of vertices.\par yeast.paj Pajek project file with complete dataset.\par\par \par \par complete dataset (ZIP, 134K)\par \par Background Interaction detection methods have led to the discovery of thousands\par of interactions between proteins, and discerning relevance within large-scale\par data sets is important to present-day biology. The dataset consists of\par protein-protein interaction network described and analyzed in (1) and available\par as an example in the software package - PIN (2).\par \par PIN class encoding:\par 1 - T, http://www.downhi.com/txt/PMhieofiFTRK.html 2 - M, 3 - U, 4 - C, 5 - F, 6 - P, 7 - G, 8 - D, 9 - O,\par 10 - E, 11 - R, 12 - B, 13 - A.\par \par yeastS.net X interacts with Y relation, short names. yeastL.net X interacts\par with Y relation, long labels. yeast.clu PIN class partition of vertices, see\par encoding. yeast.paj Pajek project file with complete dataset.\par \par References\par \par 1. Shiwei Sun, Lunjiang Ling, Nan Zhang, Guojie Li and Runsheng Chen:\par Topological structure analysis of the protein-protein interaction network in\par budding yeast. Nucleic Acids Research, 2003, Vol. 31, No. 9 2443-2450 (PDF).\par \par 2. Software package Protein Interaction Network PIN\par \par Transformed in Pajek format by V. Batagelj, 25. July 2003\par\par \par ================================================================================\par ==> GD problems <==\par ================================================================================\par \par These graphs are from the Graph Drawing 1995-2006 Contests.\par \par GD97_b:\par NOTE regarding conversion for UF sparse matrix collection: in original the data\par every edge appears exactly twice, with the same edge weight. It could be a \par multigraph, but it looks more like a graph. The duplicate edges are removed in\par this version. You can always add them back in yourself; just look at 2*A. \par \par }{ \rtlch\fcs1 \af0 \ltrch\fcs0 \fs24\insrsid6493368\charrsid1074055 \par }\pard \ltrpar\qj \li0\ri0\sl180\slmult0\nowidctlpar\wrapdefault\aspalpha\aspnum\faauto\adjustright\rin0\lin0\itap0\pararsid6493368 {\rtlch\fcs1 \af0 \ltrch\fcs0 \insrsid1074055 \par \par \par }{\rtlch\fcs1 \af0\afs30 \ltrch\fcs0 \fs30\cf17\dbch\af18\insrsid1074055\charrsid1074055 \hich\af0\dbch\af18\loch\f0 Free Document Search Engine. support all pdf,DOC,PPT,RTF,XLS,TXT\hich\af0\dbch\af18\loch\f0 ,Ebook! \hich\af0\dbch\af18\loch\f0 F \hich\af0\dbch\af18\loch\f0 ree\hich\af0\dbch\af18\loch\f0 \hich\af0\dbch\af18\loch\f0 download! You can search all kind of documents!}{\rtlch\fcs1 \af0\afs30 \ltrch\fcs0 \fs30\cf17\dbch\af18\insrsid6493368\charrsid1074055 \hich\af0\dbch\af18\loch\f0 }{ \rtlch\fcs1 \af0\afs30 \ltrch\fcs0 \fs30\cf17\insrsid1074055\charrsid1074055 \par }{\field\fldedit{\*\fldinst {\rtlch\fcs1 \af0\afs28 \ltrch\fcs0 \fs28\cf11\insrsid14892288\charrsid1074055 \hich\af0\dbch\af13\loch\f0 HYPERLINK "http://www.downhi.com/"}{\rtlch\fcs1 \af0\afs28 \ltrch\fcs0 \fs28\cf11\insrsid10707375\charrsid1074055 {\*\datafield 00d0c9ea79f9bace118c8200aa004ba90b0200000003000000e0c9ea79f9bace118c8200aa004ba90b4e00000068007400740070003a002f002f00770065006e00640061006e0067002e0064006f00630073006f0075002e0063006f006d002f000000795881f43b1d7f48af2c825dc485276300000000a5ab0000000000}} }{\fldrslt {\rtlch\fcs1 \af0\afs28 \ltrch\fcs0 \cs17\fs28\ul\cf2\insrsid14892288\charrsid1074055 \hich\af0\dbch\af13\loch\f0 http://www.downhi.com/}}}\sectd \linex0\headery851\footery992\colsx425\endnhere\sectlinegrid312\sectspecifyl\sectrsid6493368\sftnbj {\rtlch\fcs1 \af0\afs28 \ltrch\fcs0 \fs28\cf11\insrsid6493368\charrsid1074055 \par }{\rtlch\fcs1 \af0\afs28 \ltrch\fcs0 \fs28\cf11\insrsid15098623\charrsid1074055 \par }}