Learning Interestingness Measures in Terminology Extraction. A ROC-based approach.pdfVIP

下载本文档

7
0
约 8页
2017-04-13 发布于江苏
举报
版权申诉

Learning Interestingness Measures in Terminology Extraction. A ROC-based approach.pdf

1、本文档共8页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
5、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
6、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
7、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
8、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Learning Interestingness Measures in Terminology Extraction. A ROC-based approach

Learning Interestingness Measures in Terminology Extraction A ROC-based approach Mathieu Roche and Je?ro?me Aze? and Yves Kodratoff and Miche?le Sebag Abstract. In the field of Text Mining, a key phase in data prepa- ration is concerned with the extraction of terms, i.e. collocation of words attached to specific concepts (e.g. Philosophy-Dissertation). In this paper, Term Extraction is formalized as a supervised learning task, extracting a ranking hypothesis from a set of terms labeled as relevant/irrelevant by the expert. This task is tackled using the evolu- tionary algorithm ROGER, optimizing the area under the ROC curve attached to a ranking hypothesis. Empirical validation on two real-world applications demonstrates outstanding improvements compared to state-of-art interestingness measures in Term Extraction. The approach is found robust across domains (Molecular Biology, Curriculum Vit?) and languages (En- glish, French). 1 INTRODUCTION Besides the known difficulties of data mining [19], text mining presents specific difficulties due to the structure of documents and natural language [15, 20]. In particular, the construction of ontologies or terminologies [4, 22], a central task in text mining, aims at control- ling the polysemy and synonymy phenomenons through structuring the words and their meanings in the application domain. A preliminary for ontology construction is to extract the domain terms, or word collocations [4, 18, 22, 30]. Indeed, the meaning of a term (e.g. attribute-value representation) is not related to the mean- ing of its components in a simple way. Therefore, terms must be ex- tracted to enable the conceptual analysis of the corpus documents. Term extraction involves two tasks: detecting “interesting” col- location of words (candidate terms); classifying them according to classes predefined by the expert. This paper focuses on the detection of interesting terms, and more precisely on defining an interestingness measure on the word collo-