As I was writing about Ian Eslick‘s talk in the International Lisp Conference ’09 I have found out projects that he was connected to in the field of reasoning over semantic networks and common sense reasoning:
– ConceptNet aims to give computers access to common-sense knowledge, the kind of information that ordinary people know but usually leave unstated.
The data in ConceptNet is being collected from ordinary people who contributed it over the Web. ConceptNet represents this data in the form of a semantic network, and makes it available to be used in natural language processing and intelligent user interfaces.
– Commonsense Computing Initiative at the MIT Media Lab: When people communicate, they rely on a large body of shared common sense knowledge in order to understand each other. Many barriers we face today in artificial intelligence and user interface design are due to the fact that computers do not share this knowledge. To improve computers’ understanding of the world that people live in and talk about, we need to provide them with usable knowledge about the basic relationships between things that nearly every person knows.
– Open Mind Common Sense: enables computers to learn general knowledge from ordinary people over the Web. It is the product of ongoing research by the Commonsense Computing Initiative at the MIT Media Lab. You can find related projects that other groups are working on at OpenMind.org. The information collected by this site becomes part of ConceptNet, an open-source, multilingual semantic network of general knowledge. The reasoning is powered by Divisi, a tool for reasoning by analogy over semantic networks.
– Divisi: a library for reasoning by analogy and association over semantic networks, including common sense knowledge.
Divisi uses a sparse higher-order SVD can help find related concepts, features, and relation types in any knowledge base that can be represented as a semantic network. By including common sense knowledge from ConceptNet, the results can include relationships not expressed in the original data but related by common sense.
It is a library written in Python, using a C library (SVDLIBC) to perform the sparse SVD operation using the Lanczos algorithm. Other mathematical computations are performed by NumPy.