Efforts to automatically acquire world knowledge from text suffer from the lack of an easy means of evaluating the resulting knowledge. We describe initial experiments using Mechanical Turk to crowdsource evaluation to nonexperts for little cost, resulting in a collection of factoids with associated quality judgements. We describe the method of acquiring usable judgements from the public and th...

Commonsense reasoning and probabilistic planning are two of the most important research areas in artificial intelligence. This paper focuses on Integrated commonsense Reasoning and probabilistic Planning (IRP) problems. On one hand, commonsense reasoning algorithms aim at drawing conclusions using structured knowledge that is typically provided in a declarative way. On the other hand, probabili...

Applications are increasingly expected to make smart decisions based on what humans consider basic commonsense. An often overlooked but essential form of commonsense involves comparisons, e.g. the fact that bears are typically more dangerous than dogs, that tables are heavier than chairs, or that ice is colder than water. In this paper, we first rely on open information extraction methods to ob...

In an enterprise called "deep lexical semantics", we develop various core theories of fundamental commonsense phenomena and define English word senses by means of axioms using predicates explicated in these theories. This enables deep inferences that require commonsense knowledge about how the world functions. There are difficulties in our approach to manually axiomatize words and commonsense k...

Understanding language requires both linguistic knowledge and knowledge about how the world works, also known as common-sense knowledge. We attempt to characterize the kinds of common-sense knowledge most often involved in recognizing textual entailments. We identify 20 categories of common-sense knowledge that are prevalent in textual entailment, many of which have received scarce attention fr...

When teachers want to introduce a discussion about the use of cigarette to a group of teenagers, it can be interesting to know what this group thinks about it in order to contextualize the approach to their profile. It is proposed here a framework to instance web game supported by common sense knowledge to approach the called “transversal themes” of the school curriculum, like healthcare, sexua...

TACITUS is a text understanding system being developed at SRI In­ ternational. One of the main components in the system is a knowledge­ base which contains commonsense and domain specific world knowledge encoded as axioms in a first order predicate calculus language. The prime function o f the knowledgebase is to provide extra-linguistic facts to be used in the resolution of a range of ambiguit...

We consider the problem of how enormous databases of “common sense” knowledge can be both learned and utilized in reasoning in a computationally efficient manner. We propose that this is possible if the learning only occurs implicitly, i.e., without generating an explicit representation. We show that it is feasible to invoke such implicitly learned knowledge in essentially all natural tractable...

The discussion of whether the role information encoded by verbs is anticipatorily used in sentence comprehension apart from the world knowledge in the context has been a while. However, in head-initial languages like English, it is difficult to tell apart the use of lexical information like verbs from that of world knowledge information because roles associated with role fillers are mostly enco...

However, it is a general feature of metaphors that they violate hard or soft semantic constraints. They can be analyzed by detecting that these constraints have been violated, and describing how. Some of the gap in understanding can be closed simply by having a crisp model of argument types for the predicates that are denoted in text – a violation of those types is evidence for a metaphor. Clos...