Unsupervised Learning of Semantic Orientation from a Hundred-Billion-Word Corpus

Simple item page
dc.creatorTurney, Peter D.
dc.creatorLittman, Michael L.
dc.date2002-12-08
dc.date.accessioned2026-07-07T03:19:13Z
dc.date.available2026-07-07T03:19:13Z
dc.descriptionThe evaluative character of a word is called its semantic orientation. A positive semantic orientation implies desirability (e.g., "honest", "intrepid") and a negative semantic orientation implies undesirability (e.g., "disturbing", "superfluous"). This paper introduces a simple algorithm for unsupervised learning of semantic orientation from extremely large corpora. The method involves issuing queries to a Web search engine and using pointwise mutual information to analyse the results. The algorithm is empirically evaluated using a training corpus of approximately one hundred billion words -- the subset of the Web that is indexed by the chosen search engine. Tested with 3,596 words (1,614 positive and 1,982 negative), the algorithm attains an accuracy of 80%. The 3,596 test words include adjectives, adverbs, nouns, and verbs. The accuracy is comparable with the results achieved by Hatzivassiloglou and McKeown (1997), using a complex four-stage supervised learning algorithm that is restricted to determining the semantic orientation of adjectives.
dc.description11 pages, issued 2002
dc.identifierhttps://arxiv.org/abs/cs/0212012
dc.identifierhttp://arxiv.org/abs/cs/0212012
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/31374
dc.subjectMachine Learning
dc.subjectInformation Retrieval
dc.subjectH.3.1; H.3.3; I.2.6; I.2.7
dc.titleUnsupervised Learning of Semantic Orientation from a Hundred-Billion-Word Corpus
dc.typetext

Files

Collections

arXiv