Learning to plausibly reason with minimal user intervention could significantly improve knowledge acquisition. We describe how to integrate graph-based heuristic generalization, higher-order knowledge, and reinforcement learning to learn to produce plausible inferences with only small amounts of user training. Experiments on ResearchCyc KB contents show significant improvement in Q/A performance with high accuracy.

In recent years, ConceptNet has gained notoriety in the Natural Language Processing (NLP) as a textual commonsense knowledge base (CSKB) for its utilization of k-lines (Liu and Sing, 2004a) which make it suitable for making practical inferences on corpora (Liu and Sing, 2004b). However, until now, ConceptNet has lacked support for many non-English languages. To alleviate this problem, we have implemented a software toolset for the Japanese Language that allows Japanese to be used with ConceptNet's concept inference system. This paper discusses the implementation of this toolset and a possible path for the development of toolsets in other languages with similar features.

This paper presents our work in developing a commonsense knowledge source based on semantic concepts about objects, activities and their relationships in a child’s daily life. This commonsense ontology is then used by our automatic story generator to output children's stories of the fable form from a given input picture. The generated story is a narration of the events of a basic plot that flows from negative to positive (rule violation to value acquisition), using themes that are familiar to children. The paper ends with descriptions of further investigations that are underway to extend the system, including using a formal upper ontology to represent storytelling knowledge, and the generation of stories from a given set of sequential scenes.

Games with A Purpose (GWAP) has been demonstrated to be efficient in collecting large amount of knowledge from online users, e.g. Verbosity and Virtual Pet game. However, its effectiveness in knowledge base (KB) construction has not been explored in previous research. This paper examines the knowledge collected in the Vir- tual Pet game and presents an approach to collect more knowledge driven by the existing relations in KB. In this paper, goal-oriented knowledge collection successfully draws 10572 answers for the "food” domain. The answers are verified by online voting to show that 92.07% of them are good sentences and 95.89% of them are new sentences. This result is a significant improvement over the original Virtual Pet game, with 80.58% good sentences and 67.56% weekly new information.

We present CrossBridge, a practical algorithm for retrieving analogies in large, sparse semantic networks. Other algorithms adopt a generate-and-test approach, retrieving candidate analogies by superficial similarity of concepts, then testing them for the particular relations involved in the analogy. CrossBridge adopts a global approach. It organizes the entire knowledge space at once, as a matrix of small concept-and-relation subgraph patterns versus actual occurrences of subgraphs from the knowledge base. It uses the familiar mathematics of dimensionality reduction to reorganize this space along dimensions representing approximate semantic similarity of these subgraphs. Analogies can then be retrieved by simple nearest-neighbor comparison. CrossBridge also takes into account not only knowledge directly related to the source and target domains, but also a large background Commonsense knowledge base. Commonsense influences the mapping between domains, preserving important relations while ignoring others. This property allows CrossBridge to find more intuitive and extensible analogies. We compare our approach with an implementation of structure mapping and show that our algorithm consistently finds analogies in cases where structure mapping fails. We also present some discovered analogies.

Common sense collection has long been an important subfield of AI. This paper introduces a combined architecture for commonsense harvesting by text mining and a game with a purpose. The text miner module uses a seed set of known facts (sampled from ConceptNet) as training data and produces candidate commonsense facts mined from corpora. The game module taps humans' knowledge about the world by letting them play a simple slot-machine-like game. The proposed system allows us to collect significantly better commonsense facts than the state-of-the-art text miner alone, as shown experimentally for 5 rather different types of commonsense relations.

Coarse word sense disambiguation (WSD) is an NLP task that is both important and practical: it aims to distinguish senses of a word that have very different meanings, while avoiding the complexity that comes from trying to finely distinguish every possible word sense. Reasoning techniques that make use of common sense information can help to solve the WSD problem by taking word meaning and context into account. We have created a system for coarse word sense disambiguation using blending, a common sense reasoning technique, to combine information from SemCor, WordNet, ConceptNet and Extended WordNet. Within that space, a correct sense is suggested based on the similarity of the ambiguous word to each of its possible word senses. The general blending-based system performed well at the task, achieving an f-score of 80.8\% on the 2007 SemEval Coarse Word Sense Disambiguation task.

Colorizer is a program that hypothesizes color values that represent a given word or sentence, taking into account both physical descriptions of objects and their emotional connotations. This new application of common sense reasoning uses background knowledge about the world to build a model of the connections between everyday things, and uses this model to guess an appropriate color for a word. Colorizer can run over either static text or real time input, such as a speech recognition stream. It has applications in games, the arts, and webpage design.

What commonsense knowledge do intelligent systems need, in order to recover from failures or deal with unexpected situations? It is impractical to represent predetermined solutions to deal with every unanticipated situation or provide predetermined fixes for all the different ways in which systems may fail. We contend that intelligent systems require only a finite set of anomaly-handling strategies to muddle through anomalous situations. We describe a generalized metacognition module that implements such a set of anomaly-handling strategies and that in principle can be attached to any host system to improve the robustness of that system. Several implemented studies are reported, that support our contention.

The KNEXT system produces a large volume of factoids from text, expressing possibilistic general claims such as that 'A PERSON MAY HAVE A HEAD' or 'PEOPLE MAY SAY SOMETHING'. We present a rule-based method to sharpen certain classes of factoids into stronger, quantified claims such as 'ALL OR MOST PERSONS HAVE A HEAD' or 'ALL OR MOST PERSONS AT LEAST OCCASIONALLY SAY SOMETHING' -- statements strong enough to be used for inference. The judgement of whether and how to sharpen a factoid depends on the semantic categories of the terms involved and the strength of the quantifier depends on how strongly the subject is associated with what is predicated of it. We provide an initial assessment of the quality of such automatic strengthening of knowledge and examples of reasoning with multiple sharpened premises.