Commonsense Reasoning in and Over Natural Language [chapter]

Hugo Liu, Push Singh
<span title="">2004</span> <i title="Springer Berlin Heidelberg"> <a target="_blank" rel="noopener" href="" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
ConceptNet is a very large semantic network of commonsense knowledge suitable for making various kinds of practical inferences over text. ConceptNet captures a wide range of commonsense concepts and relations like those in Cyc, while its simple semantic network structure lends it an ease-of-use comparable to WordNet. To meet the dual challenge of having to encode complex higher-order concepts, and maintaining ease-of-use, we introduce a novel use of semi-structured natural language fragments as
more &raquo; ... the knowledge representation of commonsense concepts. In this paper, we present a methodology for reasoning flexibly about these semi-structured natural language fragments. We also examine the tradeoffs associated with representing commonsense knowledge in formal logic versus in natural language. We conclude that the flexibility of natural language makes it a highly suitable representation for achieving practical inferences over text, such as context finding, inference chaining, and conceptual analogy. What is ConceptNet? ConceptNet ( is the largest freely available, machine-useable commonsense resource. Structured as a network of semi-structured natural language fragments, ConceptNet presently consists of over 250,000 elements of commonsense knowledge. We were inspired dually by the range of commonsense concepts and relations in Cyc (Lenat, 1995) , and by the ease-of-use of WordNet (Fellbaum, 1998) , and hoped to combine the best of both worlds. As a result, we adopted the semantic network representation of WordNet, but extended the representation in several key ways. First, we extended WordNet's lexical notion of nodes to a conceptual notion of nodes, but we kept the nodes in natural language, because one of the primary strengths of WordNet in the textual domain is that its knowledge representation is itself textual. ConceptNet's nodes are thus natural language fragments which are semi-structured according to an ontology of allowable syntactic patterns, and accommodate both first-order concepts given as noun phrases (e.g. "potato chips"), and second-order concepts given as verb phrases (e.g. "buy potato chips"). Second, we extended WordNet's small ontology of semantic relations, which are primarily taxonomic in nature, to include a richer set of relations appropriate to concept-level nodes. At present there are 19 semantic relations used in ConceptNet, representing categories of, inter alia, temporal, spatial, causal, and functional
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="">doi:10.1007/978-3-540-30134-9_40</a> <a target="_blank" rel="external noopener" href="">fatcat:tz5zh7tohbgbjlyyk4jyveyq3q</a> </span>
<a target="_blank" rel="noopener" href="" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href=""> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> </button> </a>