A Google+ dataset of 211,187 users, 1,506,896 circle memberships, collected by Fire et al. via Google+ API (2013). Each record contains directed edges between Google+ users representing circle memberships (user → circle). Google+ social graph capturing directed circle relationships, where users add others to their circles, creating a directed social graph structure. Suitable for link prediction, social network analysis, community detection, circle-based social behavior analysis.
A BibSonomy academic social bookmarking platform dataset of 210,467 nodes and 453,987 edges, collected by KONECT via BibSonomy API. Each record contains directed: False, weighted: True, temporal: True. Temporal bipartite network of BibSonomy users and publications they bookmarked, with Unix timestamps.. Suitable for social bookmarking analysis, academic recommendation, temporal bipartite networks.
A GitHub (user-repository starring/watching network) dataset of 177,386 nodes and 440,237 edges, collected by KONECT via GitHub API. Each record contains directed: False, weighted: False, temporal: False. Bipartite network of GitHub users and repositories — edges represent starring or watching relationships.. Suitable for open source collaboration analysis, bipartite network analysis, software ecosystem research.
A Last.fm music streaming and social platform dataset of 174,077 nodes and 894,472 edges, collected by KONECT via Last.fm API. Each record contains directed: False, weighted: True, temporal: True. Temporal bipartite network of Last.fm users and bands they listened to, with Unix timestamps per play event.. Suitable for music recommendation, temporal bipartite networks, listening behavior analysis.
A DBTropes (structured TV Tropes database) dataset of 152,093 nodes and 3,232,134 edges, collected by M. Kiesel et al. via TV Tropes crawl. Each record contains directed: False, weighted: False. Bipartite network of films and narrative tropes they contain from DBTropes — useful for computational narrative analysis.. Suitable for computational narrative analysis, bipartite network analysis, film studies.
A Petster.com (pet social network — cats) dataset of 149,700 nodes and 5,449,275 edges, collected by KONECT via Petster crawl. Each record contains directed: False, weighted: False, signed: False. Friendship network of cat profiles on a pet social networking site.. Suitable for social network analysis, community detection.
A Princeton WordNet lexical database dataset of 146,005 nodes and 656,999 edges, collected by Princeton University. Each record contains directed: False, weighted: False. Semantic network of English words linked by synonymy, hypernymy, and other lexical relations from WordNet.. Suitable for natural language processing, semantic network analysis, word embedding evaluation.
A Prosper.com (peer-to-peer lending platform) dataset of 89,269 nodes and 3,330,225 edges, collected by KONECT via Prosper API. Each record contains directed: True, weighted: True, signed: False, temporal: True. Directed trust network from a peer-to-peer lending platform where users vouch for borrowers. Temporal edges with Unix timestamps enable longitudinal trust propagation analysis.. Suitable for trust network analysis, temporal network analysis, peer-to-peer lending.
A Jester Online Joke Recommender System (UC Berkeley) dataset of 73,521 nodes and 4,136,360 edges, collected by K. Goldberg et al. via Jester system. Each record contains directed: False, weighted: True, temporal: False. Bipartite user-joke rating network with continuous ratings from -10 to +10 — one of the earliest large-scale collaborative filtering datasets.. Suitable for recommender systems, collaborative filtering, continuous rating prediction.
A TREC web corpus dataset of 58,434 nodes and 978,446 edges, collected by T. Gottron et al.. Each record contains directed: False, weighted: True. Bipartite network of web documents and terms from the TREC web corpus.. Suitable for document-term network analysis, information retrieval, web graph research.
A Slashdot (technology news discussion forum) dataset of 51,083 nodes and 131,175 edges, collected by KONECT via Slashdot crawl. Each record contains directed: True, weighted: True, temporal: True. Directed reply network of Slashdot discussion threads — different from the Slashdot Zoo signed network, this captures comment reply structure.. Suitable for discussion thread analysis, online community structure, reply network analysis.
A Twitter dataset of 38,918 users and 32,523 directed, weighted reply edges (36,902 replies in all), collected by De Domenico et al. via the Twitter Streaming API around the Higgs boson announcement, 1-7 July 2012. Each record holds the replying user ID, the replied-to user ID and the number of replies between them. Complements the Higgs retweet and mention networks from the same collection. Suitable for information diffusion, scientific communication on social media and reply network structure.
A Academic co-authorship bipartite network dataset of 38,741 nodes and 58,595 edges, collected by T. Opsahl. Each record contains directed: False, weighted: False, signed: False. Large bipartite collaboration network of authors and papers — useful for studying academic collaboration patterns.. Suitable for bipartite network analysis, academic collaboration, community detection.
A MySpace dataset of 80,749 messages between 36,459 users (41,533 directed user pairs) over two months, collected by Šuvakov et al. (Dryad, 2013). Each record holds the sender, the recipient, the time in minutes since the start of the window, and the message's SentiStrength positive (1 to 5) and negative (-1 to -5) scores. Suitable for sentiment dynamics, dialogue patterns and temporal network analysis.
A Internet router topology dataset of 34,761 nodes and 107,720 edges, collected by KONECT via traceroute measurements. Each record contains directed: False, weighted: True. Temporal internet router topology network with Unix timestamps per measurement.. Suitable for internet topology analysis, network robustness, temporal network analysis.
A Linux kernel source code (function call graph) dataset of 30,837 nodes and 213,954 edges, collected by KONECT via static analysis of Linux kernel. Each record contains directed: True, weighted: False. Directed call graph of the Linux kernel — nodes are functions, edges are function calls.. Suitable for software network analysis, dependency network research, community detection.
Contacts between people measured by small wireless devices they carried, from the Haggle project (Chaintreau et al.), as distributed by KONECT (contact): 28,244 timestamped contacts among 274 people, forming 2,124 distinct pairs. An edge means two people's devices were within range of each other; timestamps are relative and the records span about four days. Useful for opportunistic networking, temporal network analysis and epidemic modelling on human contact data.
The co-authorship network of the arXiv High Energy Physics – Phenomenology (hep-ph) section, as distributed by KONECT (ca-cit-HepPh), from SNAP's arXiv data: 28,093 authors linked by 4,596,803 collaboration records, one for each pair of authors on a paper, published between March 1992 and March 2002. The records form 3,148,447 distinct co-author pairs, each timestamped with its paper's date. Useful for studying scientific collaboration, the growth and densification of networks over time, and community structure.
A Linux Kernel Mailing List (LKML) — open source development community dataset of 27,927 nodes and 242,976 edges, collected by Ingo Scholtes et al. via mailing list archive. Each record contains directed: True, weighted: True, signed: False, temporal: True. Temporal directed reply network of the Linux Kernel Mailing List — one of the longest-running and most active open source development communities. Edges carry Unix timestamps enabling longitudinal analysis.. Suitable for temporal network analysis, community detection, open source collaboration.
A arXiv High Energy Physics — Theory (1991–2003) dataset of 27,770 nodes and 352,807 edges, collected by J. Leskovec et al. via arXiv. Each record contains directed: True, weighted: False. Citation network of HEP-TH papers on arXiv — directed, covers theoretical physics literature.. Suitable for citation network analysis, academic influence, graph mining.
A Internet autonomous system topology (CAIDA, 2007) dataset of 26,475 nodes and 53,381 edges, collected by CAIDA via BGP routing data. Each record contains directed: False, weighted: False. Internet autonomous system (AS) peering network — nodes are AS networks, edges are peering relationships.. Suitable for internet topology analysis, network robustness, community detection.
Bipartite user–page edit networks from Wikimedia dumps, as distributed by KONECT; each edge is one edit event with its Unix timestamp. The package holds two separate networks: the French Wikinews (edit-frwikinews: 1,408 users, 25,039 pages, 193,618 edits) and the English Wikipedia (edit-enwiki, 1.75 GB: 8,116,897 users, 42,640,545 pages, 572,591,272 edits per KONECT). The node and edge counts and the analytics on this page were computed from the French Wikinews network only. Suitable for wiki collaboration analysis, editor behavior, temporal bipartite networks.
A Cora machine learning paper database dataset of 23,166 nodes and 91,500 edges, collected by McCallum et al.. Each record contains directed: True, weighted: False. Classic citation network of machine learning papers — one of the most used benchmarks for graph neural network node classification.. Suitable for graph neural networks, node classification, citation network analysis.
A Edinburgh Associative Thesaurus (word association study) dataset of 23,132 nodes and 312,310 edges, collected by G.R. Kiss et al.. Each record contains directed: True, weighted: True. Directed word association network from the Edinburgh Associative Thesaurus — edges connect stimulus words to response words.. Suitable for semantic network analysis, word association modeling, NLP.
A arXiv Astrophysics (collaboration network 1992–2003) dataset of 18,771 nodes and 198,050 edges, collected by J. Leskovec et al. via arXiv. Each record contains directed: False, weighted: False. Co-authorship network of astrophysics papers on arXiv — nodes are authors, edges connect co-authors.. Suitable for collaboration network analysis, community detection, academic social networks.
Yelp coffee-shop reviews for computational branding analytics (Wang, Lin & Kominek, EMNLP 2013): 17,894 reviews in train (10,495), dev (3,617) and test (3,782) splits. Each row holds the brand (Starbucks, Dunkin', other), satisfaction (negative, positive, neutral), the reviewer's gender (missing as '?' in some rows), US region, and the review text. There are no review, business or date fields. Research use only, per the README.
A Escort review website (sex worker review community) dataset of 16,730 nodes and 39,044 edges, collected by Rocha et al. via web crawl. Each record contains directed: False, weighted: True, signed: True, temporal: True. Bipartite signed network of escort reviews: clients rate escorts +1 (positive), 0 (neutral), or -1 (negative), with Unix timestamps. One of the few large-scale signed bipartite networks available.. Suitable for signed network analysis, bipartite network analysis, trust propagation.
A Google web graph subset dataset of 15,763 nodes and 171,206 edges, collected by KONECT. Each record contains directed: True, weighted: False. Web hyperlink graph used with the CFinder community detection tool.. Suitable for community detection, web graph analysis.
A Free On-line Dictionary of Computing (FOLDOC) dataset of 13,356 nodes and 120,238 edges, collected by Denis Howe. Each record contains directed: True, weighted: True. Directed network of computing terms linked by cross-references in the FOLDOC dictionary.. Suitable for knowledge network analysis, dictionary network structure, NLP.
A LASAGNE online social network dataset dataset of 12,643 nodes and 57,453 edges, collected by KONECT. Each record contains directed: True, weighted: True. Directed network from the LASAGNE project — a small online social network dataset.. Suitable for social network analysis, community detection.
A DBLP Computer Science Bibliography dataset of 12,591 nodes and 49,743 edges, collected by KONECT via DBLP database. Each record contains directed: True, weighted: False. Directed citation network of computer science papers in the DBLP bibliography.. Suitable for citation network analysis, academic influence, community detection.
A PGP (Pretty Good Privacy) key signing network dataset of 10,680 nodes and 24,316 edges, collected by M. Boguñá et al.. Each record contains directed: False, weighted: False, signed: False. Web of trust network from PGP key signings — nodes are PGP users, edges are key signatures.. Suitable for trust network analysis, community detection, graph algorithm benchmarking.
A Chess game records (professional matches) dataset of 7,301 nodes and 60,046 edges, collected by KONECT via chess game database. Each record contains directed: True, weighted: True, signed: True. Directed network of chess game outcomes: edges go from winner to loser, with timestamps. Draws encoded as 0 weight.. Suitable for competition network analysis, ranking algorithms, temporal network analysis.
A Reactome biological pathway database dataset of 6,327 nodes and 147,547 edges, collected by Reactome consortium. Each record contains directed: False, weighted: False, signed: False. Protein-protein interaction network from the Reactome pathway database — large-scale biological network.. Suitable for biological network analysis, community detection, protein interaction research.
A US Western States power grid infrastructure dataset of 4,941 nodes and 6,594 edges, collected by D.J. Watts & S.H. Strogatz. Each record contains directed: False, weighted: False, signed: False. Classic small-world network of the US western power grid — nodes are generators/transformers, edges are transmission lines.. Suitable for infrastructure network analysis, small-world networks, robustness analysis.
A OpenFlights.org airport route database dataset of 3,425 nodes and 37,595 edges, collected by OpenFlights.org. Each record contains directed: True, weighted: True. Directed network of airports connected by flight routes — nodes are airports, edges are direct routes.. Suitable for transportation network analysis, network robustness, hub detection.
A Human protein interaction network (Vidal lab) dataset of 3,133 nodes and 6,726 edges, collected by M. Vidal et al.. Each record contains directed: False, weighted: False. Human protein-protein interaction network from the Vidal lab — classic biological network benchmark.. Suitable for biological network analysis, protein interaction research.
An adolescent friendship network from the Add Health study (2,539 students, 12,969 edges), compiled by James Moody. Each record contains directed edges: True, weighted: True, signed: False. This directed network was created from a survey that took place in 1994/1995. Each student was asked to list his 5 best female and his 5 male friends. A node represents a student and an edge between two students shows that the left student chose the right student as a friend. Higher edge weights ind. Suitable for social network analysis, community detection.
A Human protein interaction network (Figeys lab) dataset of 2,239 nodes and 6,452 edges, collected by D. Figeys et al.. Each record contains directed: True, weighted: False. Human protein-protein interaction network from the Figeys lab.. Suitable for biological network analysis, protein interaction research.
A UC Irvine online messaging system dataset of 1,899 nodes and 20,296 edges, collected by T. Opsahl. Each record contains directed: True, weighted: True, signed: False. Temporal directed messaging network among UC Irvine students with Unix timestamps.. Suitable for temporal network analysis, online social network analysis. The file holds 59,835 messages, one row per message; the edge count of 20,296 counts distinct pairs.
A Protein-protein interaction network dataset of 1,870 nodes and 2,277 edges, collected by KONECT. Each record contains directed edges: False, weighted: False, signed: False. This undirected network contains protein interactions contained in yeast. Research showed that proteins with a high degree were more important for the surivial of the yeast than others. A node represents a protein and an edge represents a metabolic interaction between two proteins. The network cont. Suitable for biological network analysis, community detection.
A Petster.com (pet social network — hamsters) dataset of 1,858 nodes and 12,534 edges, collected by KONECT via Petster crawl. Each record contains directed: False, weighted: False, signed: False. Friendship network of hamster profiles on a pet social networking site.. Suitable for social network analysis, community detection.
A Name association network dataset of 1,773 nodes and 9,131 edges, collected by J.L. Moreno. Each record contains directed edges: False, weighted: True, signed: False. This undirected network contains nouns (places and names) of the King James Version of the Bible and information about their co-occurrences. A node represents one of the above noun types and an edge indicates that two nouns appeared together in the same Bible verse. The edge multiplicity denotes ho. Suitable for semantic network analysis.
A Human protein interaction network (Stelzl et al.) dataset of 1,706 nodes and 6,207 edges, collected by U. Stelzl et al.. Each record contains directed: True, weighted: False. Human protein-protein interaction network from Stelzl et al. — yeast two-hybrid screening.. Suitable for biological network analysis, protein interaction research.
A Chicago regional road network (TNTP benchmark) dataset of 1,467 nodes and 1,298 edges, collected by Transportation Networks for Research (TNTP). Each record contains directed: False, weighted: False. Road network of the Chicago metropolitan area — classic transportation network benchmark.. Suitable for transportation network analysis, routing algorithms, network flow.
A UC Irvine online student forum dataset of 1,421 nodes and 7,089 edges, collected by T. Opsahl. Each record contains directed: False, weighted: True, signed: False. Temporal bipartite network of students posting to forum threads with Unix timestamps.. Suitable for temporal bipartite network analysis, online community analysis.
A FAA air traffic control network dataset of 1,226 nodes and 2,615 edges, collected by A. Ma'ayan et al.. Each record contains directed: True, weighted: False. FAA air traffic control network from the Ma'ayan lab dataset collection.. Suitable for infrastructure network analysis, biological network benchmarking.
A Political blogosphere (US 2004 election) dataset of 1,224 nodes and 19,025 edges, collected by Adamic & Glance via blog crawl. Each record contains directed edges: True, weighted: False, signed: False. This directed network contains front-page hyperlinks between blogs in the context of the 2004 US election. A node represents a blog and an edge represents a hyperlink between two blogs.. Suitable for political polarization, community detection, information diffusion.
The international E-road network, a road network located mostly in Europe, compiled by Lovro Šubelj and Marko Bajec and distributed by KONECT (subelj_euroroad): 1,174 cities linked by 1,417 undirected roads, where an edge means two cities are joined by an E-road. City names are included. Useful for studying spatial and infrastructure networks, network robustness and community detection.
A University Rovira i Virgili (email communication network) dataset of 1,133 nodes and 5,451 edges, collected by R. Guimerà et al.. Each record contains directed: False, weighted: False, signed: False. Email communication network of a university — classic benchmark for community detection algorithms.. Suitable for community detection, email network analysis, graph algorithm benchmarking.
Google
BibSonomy
GitHub
Last.fm
Petster
The internet
Slashdot
Twitter (X)
Myspace
Wikipedia
Yelp