{"pk":27366,"title":"A Data Driven Approach for Making Analogies","subtitle":null,"abstract":"Making analogies is an important way for people to explain\nand understand new concepts. Though making analogies is\nnatural for human beings, it is not a trivial task for a dia-\nlogue agent. Making analogies requires the agent to estab-\nlish a correspondence between concepts in two different\ndomains. In this work, we explore a data-driven approach\nfor making analogies automatically. Our proposed approach\nworks with data represented as a flat graphical structure,\nwhich can either be designed manually or extracted from In-\nternet data. For a given concept from the base domain, our\nanalogy agent can automatically suggest a corresponding\nconcept from the target domain, and a set of mappings be-\ntween the relationships each concept has as supporting evi-\ndence. We demonstrate the working of this algorithm by\nboth reproducing a classical example of analogy inference\nand making analogies in new domains generated from\nDBPedia data.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[{"word":"creativity; analogy; intelligent agents"}],"section":"Posters: Papers","is_remote":true,"remote_url":"https://escholarship.org/uc/item/2dx8b228","frozenauthors":[{"first_name":"Mei","middle_name":"","last_name":"Si","name_suffix":"","institution":"Rensselaer Polytechnic Institute","department":""},{"first_name":"Craig","middle_name":"","last_name":"Carlson","name_suffix":"","institution":"Rensselaer Polytechnic Institute","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"2017-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/27366/galley/17002/download/"}]}