{"id":555,"date":"2011-04-22T22:22:49","date_gmt":"2011-04-22T13:22:49","guid":{"rendered":"http:\/\/lr-www.pi.titech.ac.jp\/?p=555"},"modified":"2011-04-23T07:30:33","modified_gmt":"2011-04-22T22:30:33","slug":"%e8%bc%aa%e8%ac%9b-2","status":"publish","type":"post","link":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/?p=555","title":{"rendered":"\u8f2a\u8b1b"},"content":{"rendered":"<p>\u677e\u7530\u3055\u3093\u306e\u8f2a\u8b1b\u3067\u3059\u3002<\/p>\n<p>Ioannis P. Klapaftis and Suresh Manandhar<br \/>\nWord Sense Induction & Disambiguation Using Hierarchical Random Graphs<br \/>\n(EMNLP 2010)<br \/>\nhttp:\/\/www.aclweb.org\/anthology\/D\/D10\/D10-1073.pdf<\/p>\n<p>Abstract:<\/p>\n<p>Graph-based methods have gained attention in many areas of Natural Language Processing (NLP) including Word Sense Disambiguation (WSD), text summarization, keyword extraction and others.<\/p>\n<p>Most of the work in these areas formulate their problem in a graph-based setting and apply unsupervised graph clustering to obtain a set of clusters.<\/p>\n<p>Recent studies suggest that graphs often exhibit a hierarchical structure that goes beyond simple flat clustering.<\/p>\n<p>This paper presents an unsupervised method for inferring the hierarchical grouping of the senses of a polysemous word.<\/p>\n<p>The inferred hierarchical structures are applied to the problem of word sense disambiguation, where we show that our method performs significantly better than traditional graph-based methods and agglomerative clustering yielding improvements over state-of-the-art WSD systems based on sense induction.<\/p>\n<p>------<br \/>\n\u8a9e\u7fa9\u66d6\u6627\u6027\u89e3\u6d88 \u3092\u30b0\u30e9\u30d5\u3092\u7528\u3044\u3066\u89e3\u304f\u3068\u3044\u3046\u8ad6\u6587\u3067\u3059\u3002<br \/>\n\u30b0\u30e9\u30d5\u30d9\u30fc\u30b9\u306e\u8a9e\u7fa9\u66d6\u6627\u6027\u89e3\u6d88\u306f\u591a\u304f\u306e\u7814\u7a76\u304c\u3042\u308b\u306e\u3067\u3059\u304c\u3001\u3053\u306e\u8ad6\u6587\u306b\u304a\u3044\u3066\u7279\u5fb4\u7684\u306a\u306e\u306f\u3001\u4e00\u5ea6\u5171\u8d77\u95a2\u4fc2\u3092\u7528\u3044\u3066\u30b0\u30e9\u30d5\uff08\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u9593\u306e\u5171\u8d77\u30b0\u30e9\u30d5\uff09\u3092\u4f5c\u3063\u305f\u3042\u3068\u3067\u3001\u305d\u306e\u30b0\u30e9\u30d5\u3092\u5143\u306b\u5225\u306a\u30b0\u30e9\u30d5 (Hierarchical Random Graph\u3068\u3044\u3046\u4e8c\u5206\u6728)\u3092\u4f5c\u308a\u3001\u305d\u306e\u4e0a\u3067\u66d6\u6627\u6027\u306e\u89e3\u6d88\u3092\u884c\u3046\u3068\u3044\u3046\u70b9\u3067\u3059\u3002<\/p>\n<p>\u8a9e\u7fa9\u66d6\u6627\u6027\u89e3\u6d88\u306b\u304a\u3044\u3066\u5e83\u304f\u7528\u3044\u3089\u308c\u3066\u3044\u308b\u30d5\u30e9\u30c3\u30c8\u306a\u30af\u30e9\u30b9\u30bf\u30ea\u30f3\u30b0\u3067\u306f\u30ad\u30e3\u30d7\u30c1\u30e3\u30fc\u3059\u308b\u3053\u3068\u304c\u96e3\u3057\u3044\u8a9e\u7fa9\u306b\u5b58\u5728\u3059\u308b\u968e\u5c64\u95a2\u4fc2\u3092\u7528\u3044\u308b\u305f\u3081\u306b\u3001\u30d5\u30e9\u30c3\u30c8\u306a\u30b0\u30e9\u30d5\u3092\u968e\u5c64\u69cb\u9020\u3092\u8868\u73fe\u3057\u305f\u6728\u306b\u843d\u3068\u3059\u3001\u3068\u3044\u3046\u306e\u304c\u4e3b\u306a\u30dd\u30a4\u30f3\u30c8\u306b\u306a\u3063\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\u6728\u306e\u69cb\u9020\u3092\u63a8\u5b9a\u3059\u308b\u969b\u306e\u7d44\u307f\u5408\u308f\u305b\u7206\u767a\u306b\u5bfe\u51e6\u3059\u308b\u305f\u3081\u306b\u3001MCMC(\u30de\u30eb\u30b3\u30d5\u9023\u9396\u30e2\u30f3\u30c6\u30ab\u30eb\u30ed\u6cd5)\u3092\u7528\u3044\u3066\u3044\u307e\u3059\u304c\u3001MCMC\u3092\u4f7f\u3063\u305f\u4ed6\u306e\u7814\u7a76\u3068\u304f\u3089\u3079\u308b\u3068\u6bd4\u8f03\u7684\u5206\u304b\u308a\u3084\u3059\u3044\u4f7f\u3044\u65b9\u306b\u306a\u3063\u3066\u3044\u307e\u3059\u306e\u3067\u3001\u3042\u307e\u308a\u8eab\u69cb\u3048\u305a\u306b\u805e\u3044\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u677e\u7530\u3055\u3093\u306e\u8f2a\u8b1b\u3067\u3059\u3002 Ioannis P. Klapaftis and Suresh Manandhar Word Sense Induction &#038; Disambiguation Using Hierarchical  [&hellip;]<\/p>\n","protected":false},"author":6,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=\/wp\/v2\/posts\/555"}],"collection":[{"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=555"}],"version-history":[{"count":0,"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=\/wp\/v2\/posts\/555\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=555"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=555"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.lr.first.iir.isct.ac.jp\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=555"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}