Knext
—Knowledge Extraction from Text
We think that there is a largely untapped source of general knowledge in texts, lying at a level beneath the explicit assertional content. This knowledge consists of relationships implied to be possible in the world, or, under certain conditions, implied to be normal or commonplace in the world. For instance, the sentence ‘He entered the house through its open door’ suggests that it is possible for a person (or at least a male) to enter a house, that houses have doors, that doors can be open, etc. The goal of the present work is to derive such general world knowledge… Read more.
Browser
Over 16 million of the factoids discovered by Knext can be searched online with the Knext browser.
Downloads
Download the basic Knext system.
Download data – releases of knowledge bases, including some rated factoids.
Publications
The original paper describing Knext is:
- Lenhart K. Schubert: Can We Derive General World Knowledge from Texts? Proceedings of the Second International Conference on Human Language Technology Research (HLT 2002), March 24–27, San Diego, CA, pp. 94–97.
Some of the publications on Knext and related knowledge extraction work are:
2012
- Jonathan Gordon and Lenhart Schubert: Using Textual Patterns to Learn Expected Event Frequencies. Proceedings of the NAACL 2012 Workshop on Automatic Knowledge Base Construction and Web-Scale Knowledge Extraction (AKBC-WEKEX 2012).
2011
- Lenhart Schubert, Jonathan Gordon, Karl Stratos, and Adina Rubinoff: Towards Adequate Knowledge and Natural Inference. Proceedings of the AAAI 2011 Fall Symposium on Advances in Cognitive Systems.
- Jonathan Gordon and Lenhart Schubert: Discovering Commonsense Entailment Rules Implicit in Sentences. Proceedings of the EMNLP 2011 Workshop on Textual Entailment (TextInfer 2011).
2010
- Jonathan Gordon and Lenhart K. Schubert: Quantificational Sharpening of Commonsense Knowledge. Proceedngs of the AAAI 2010 Fall Symposium on Commonsense Knowledge.
- Jonathan Gordon, Benjamin Van Durme, and Lenhart K. Schubert: Learning from the Web: Extracting General World Knowledge from Noisy Text. Proceedings of the AAAI 2010 Workshop on Collaboratively-built Knowledge Sources and Artificial Intelligence (WikiAI 2010).
- Jonathan Gordon, Benjamin Van Durme, and Lenhart K. Schubert: Evaluation of Commonsense Knowledge with Mechanical Turk. Proceedings of the NAACL 2010 Workshop on Creating Speech and Language Data with Amazon’s Mechanical Turk.
2009
- Lenhart K. Schubert: From Generic Sentences to Scripts. IJCAI 2009 Workshop on Logic and the Simulation of Interaction and Reasoning (LSIR2).
- Benjamin Van Durme & Daniel Gildea: Topic Models for Corpus-centric Knowledge Generalization. Technical Report TR-946, Department of Computer Science, University of Rochester, Rochester, NY 14627, June 2009.
- Jonathan Gordon, Benjamin Van Durme, and Lenhart K. Schubert: Weblogs as a Source for Extracting General World Knowledge. Proceedings of the Fifth International Conference on Knowledge Capture (K-CAP 2009).
- Benjamin Van Durme, Phillip Michalak, and Lenhart K. Schubert: Deriving Generalized Knowledge from Corpora using WordNet Abstraction. Proceedings of EACL 2009.
2008
- Benjamin Van Durme & Lenhart K. Schubert: Open Knowledge Extraction through Compositional Language Processing. Symposium on Semantics in Systems for Text Processing (STEP ’08). Venice, Italy. September 22–24, 2008.
- Benjamin Van Durme, Ting Qian, and Lenhart K. Schubert: Class-Driven Attribute Extraction. Proceedings of COLING ’08. Manchester, UK. August 18–22, 2008.
2003
- Lenhart K. Schubert & Matthew Tong: Extracting and Evaluating General World Knowledge from the Brown Corpus. Proceedings of the HLT-NAACL Workshop on Text Meaning, pp. 7–13.
2002
- Lenhart K. Schubert: Can We Derive General World Knowledge from Texts? Proceedings of the Second International Conference on Human Language Technology Research (HLT 2002), March 24–27, San Diego, CA, pp. 94–97.
2001
- Aaron N. Kaplan & Lenhart K. Schubert: Measuring and Improving the Quality of World Knowledge Extracted from WordNet. Tech. Report 751, Dept of Computer Science, University of Rochester, Rochester, NY, May 2001.
For more, see the list of publications on Epilog, the Episodic Logic reasoning engine, and those on Lenhart Schubert’s website.
Acknowledgement
This work is currently supported by
- NSF IIS-0916599: General Knowledge Bootstrapping from Text
- NSF IIS-1016735: Adapting a Natural Logic Reasoning Platform to the Task of Entailment Inference
- Subcontract FY10-01-0297 under ONR STTR N00014-10-M-0297.
It has previously been supported under grants
- NSF IIS-0082928: Mining Text for General World Knowledge, 2000–2003
- NSF IIS-0328849: Deriving General World Knowledge from Texts by Abstraction of Logical Forms, 2003–2006
- NSF IIS-0535105: Knowledge Representation Mechanisms for Explicitly Self-Aware Agents, 2006–2009