资讯撷取与推荐技术期中报告28.pptVIP

資訊擷取與推薦技術期中報告Ch9. Thesaurus Construction ;outline;1.1 索引典(Thesaurus)定義 ;1.2 索引典結構;1.3 INSPEC thesaurus;A short extract from the 1979 INSPEC thesaurus;1.4 索引典參照款目說明;2. Features of thesauri;2.1 Coordination level;A postcoordinated thesaurus does not allow phrases. Instead, phrases are constructed while searching. The advantage is that the user need not worry about the exact ordering of the words in phrase. The disadvantage is that search precision may fall. Automatic thesaurus construction usually implies postcoordination.;Coordination level組合層次;2.2 Term Relationships詞彙間的關係;2.2a) Equivalence relationships;同義關係;2.2b) Hierarchical relationships層級關係;2.2b) 層屬關係;2.2c) Nonhierarchical relationships非層級關係;相關關係;Wang, Vandendorpe, and Evens (1985) provide an alternative classification of term relationships consisting of: (1)parts—wholes(整部關係) (2)collocation relations(排列關係) (3)paradigmatic relations(範例關係) (4)taxonomy and synonymy(分類及同義字) (5)antonymy relations(反義字);(1)Parts-wholes;(2)Collocation relations編排關係;(3)Paradigmatic relations;(4)Taxonomy and synonymy;(5) antonymy relations;2.3 Number of entries for each term;同形異義關係(homographs);2.4 Specificity of vocabulary 詞彙的明確性;2.5 Control on term frequency of class members;2.6 Normalization of vocabulary詞彙的標準化;3.1 Manual Thesaurus Construction;3.1 Manual Thesaurus Construction(continued);3.1 Manual Thesaurus Construction(continued);3.2 Automatic Thesaurus Construction;3.2 Automatic Thesaurus Construction (continued);3.2 Automatic Thesaurus Construction (continued);3.2.1 Thesaurus Construction from Texts;1 Construction of Vocabulary;Selection by Frequency of Occurrence;Selection by Discrimination Value (DV);Selection by Discrimination Value (DV)(continued);Selection by Poisson Method;Phrase Construction;Phrase Construction (continued);Choueka Procedure (continued);Choueka Procedure (continued);4.2 Similarity computations between terms;4.3 Organization of vocabulary;;Merging existing thesauri;conclusion;+B3E6H9LcOfRjUmXp!sw)z0