ConceptNet 2 is no longer maintained. New development is taking place on ConceptNet 3.

hugo :: research
"only as an æsthetic phenomenon is
existence and the world justified"

- nietzsche


emotus ponens picture

The ConceptNet Project V2.1
A Very-Large Semantic Network of Common Sense Knowledge

!NEW! ConceptNet Developer Wiki

<Overview> <Papers> <Download!> <Extensions> <Research using ConceptNet>

Play with the ConceptNet Flash Browser Now! Click here!




ConceptNet Team

hugo liu

push singh



What is ConceptNet? [top]

ConceptNet is a freely available commonsense knowledgebase and natural-language-processing toolkit which supports many practical textual-reasoning tasks over real-world documents right out-of-the-box (without additional statistical training) including

  • topic-jisting (e.g. a news article containing the concepts, “gun,” “convenience store,” “demand money” and “make getaway” might suggest the topics “robbery” and “crime”),
  • affect-sensing (e.g. this email is sad and angry),
  • analogy-making (e.g. “scissors,” “razor,” “nail clipper,” and “sword” are perhaps like a “knife” because they are all “sharp,” and can be used to “cut something”),
  • text summarization
  • contextual expansion
  • causal projection
  • cold document classification
  • and other context-oriented inferences

The ConceptNet knowledgebase is a semantic network presently available in two versions: concise (200,000 assertions) and full (1.6 million assertions). Commonsense knowledge in ConceptNet encompasses the spatial, physical, social, temporal, and psychological aspects of everyday life. Whereas similar large-scale semantic knowledgebases like Cyc and WordNet are carefully handcrafted, ConceptNet is generated automatically from the 700,000 sentences of the Open Mind Common Sense Project – a World Wide Web based collaboration with over 14,000 authors.

ConceptNet is a unique resource in that it captures a wide range of commonsense concepts and relations, such as those found in the Cyc knowledgebase, yet this knowledge is structured not as a complex and intricate logical framework, but rather as a simple, easy-to-use semantic network, like WordNet. While ConceptNet still supports many of the same applications as WordNet, such as query expansion and determining semantic similarity, its focus on concepts-rather-than-words, its more diverse relational ontology, and its emphasis on informal conceptual-connectedness over formal linguistic-rigor allow it to go beyond WordNet to make practical, context-oriented, commonsense inferences over real-world texts.

At the end of the day, we want ConceptNet to be simply useful to AI Researchers and computer enthusiasts who want to experiment with adding commonsense to make their smart robots and programs smarter. And it's working! ConceptNet is currently driving tens of new innovative research projects at MIT and elsewhere!

Papers about ConceptNet [top]:

A good overview paper of ConceptNet v2.1, Liu, H. & Singh, P. (2004) ConceptNet: A Practical Commonsense Reasoning Toolkit. BT Technology Journal, To Appear. Volume 22, forthcoming issue. Kluwer Academic Publishers. (paper)

Focusing on ConceptNet's natural language knowledge representation, Liu, H. & Singh, P. (2004). Commonsense Reasoning in and over Natural Language Proceedings of the 8th International Conference on Knowledge-Based Intelligent Information & Engineering Systems (KES'2004). Wellington, New Zealand. September 22-24. Lecture Notes in Artificial Intelligence, Springer 2004 (paper)

Download ConceptNet [top]

ConceptNet v2.1 is different from previous versions in that it is integrated and distributed with the MontyLingua natural language toolkit. Unfortunately, ConceptNet v2.1 datafiles are not backward-compatible, but previous versions may be downloaded here.

ConceptNet v2.1 is written in the Python programming language (to download Python, click here). It is distributed as a Python API and a standalone XML-RPC Server. If you wish to access ConceptNet from other programming languages such as Java, C++, or the .NET Platform, simply launch ConceptNet's XML-RPC Server (self-documenting) and then interface with ConceptNet via a simple XML-RPC client, which is available for all major programming languages. Sample client code is available in this great XML-RPC HOWTO.

Please fill out the following information to proceed to the download of ConceptNet Version 2.1 for Python and the standalone XML-RPC Server. This information will not be shared or sold with others.

What is your work/school email address?

Your company, university, or institution (+ division / group)

Would you like to tell us more about how you are planning to use ConceptNet or leave other feedback for us? (optional)

Have you downloaded ConceptNet before?
Yes   No 

If you have read and agree to these terms of use, click below to continue to the download
(your IP address will also be recorded):

(Download is a 28MB zip file)


ConceptNet Extensions [top]

Thanks to contributors from the ConceptNet community, ConceptNet is implemented in several programming languages, and we now have a variety of ConceptNet-related tools and browsers. If you would like your tool added to this list, please email us. Remember, OMCSNet=ConceptNet.

Previous Versions of ConceptNet

Extensions and Tools for Previous Versions of ConceptNet

  • RubyCon: Concept Processing Toolkit - RubyCon builds upon the work of the ConceptNet project (version 1.3.1), implementing ConceptNet’s semantic network of 280,000+ assertions and graph-processing algorithms into a set of reusable objects in the Ruby programming language.
  • OMCSNet Browser for MacOS X - A simple Cocoa-based knowledge browser for OMCSNet. The package includes source code, project files, and OMCSNet data files. It shows how to incorporate OMCSNet within an Objective-C app. Written by Edison Thomaz.
  • Web / CGI Interface to OMCSNet - Query the OMCSNetCPP Semantic Network and inference tools from a web form / cgi-bin! Written by Stuart Horner, using OMCSNetCPP.
  • OMCSNetCPP+WNLG - A wordsense disambiguated version of OMCSNet, merged with WordNet data, and annotated with part-of-speech information. Written by Elliot Turner


Research Using ConceptNet [top]
OMCSNet = ConceptNet

S. A. Inverso, N. Hawes, J. Kelleher, R. Allen, and K. Haase."Think And Spell: Context-Sensitive Predictive Text for an Ambiguous Keyboard Brain-Computer Interface Speller", to appear in the Journal of Biomedizinische Technik special issue Proceedings of the 2nd International Brain-Computer Interface Workshop and Training Course, Graz, Austria, September 16-18, 2004 (paper)

Ashwani Kumar, Sharad C. Sundararajan, Henry Lieberman (2004). Common Sense Investing: Bridging the Gap Between Expert and Novice. Conference on Human Factors in Computing Systems (CHI 04), Vienna, Austria. (paper)

Tom Stocky, Alexander Faaborg, Henry Lieberman (2004). A Commonsense Approach to Predictive Text Entry. Conference on Human Factors in Computing Systems (CHI 04), Vienna, Austria. (website)

Hugo Liu (2003). Unpacking Meaning from Words: A Context-Centered Approach to Computational Lexicon Design. CONTEXT 2003: 218-232 (paper)

Nathan Eagle, Push Singh, and Alex (Sandy) Pentland (2003). Common sense conversations: understanding casual conversation using a common sense database. Proceedings of the Artificial Intelligence, Information Access, and Mobile Computing Workshop (IJCAI 2003). Acapulco, Mexico.

Rami Musa, Madleina Scheidegger, Andrea Kulas, Yoan Anguilet (2003).
GloBuddy, a Dynamic Broad Context Phrase Book. Proceedings of the International and Interdisciplinary Conference on Modeling Context. pp. 467-474 (paper)

Austin Wang. (2002). Turning-taking in a Collaborative Storytelling Agent. Masters Thesis. MIT Department of Electrical Engineering and Computer Science.



H U G O . . L I U ...
ambient intelligence group

commonsense computing group
sign my guestbook
hugo at media dot mit dot edu