作者: Shlomo Argamon , Shlomo Levitan , Casey Whitelaw , Navendu Garg , Paul Chase
DOI: 10.1002/ASI.V58:6
关键词: Rhetoric 、 Character (computing) 、 Text processing 、 Computer science 、 Style (sociolinguistics) 、 Text mining 、 Artificial intelligence 、 Natural language processing 、 Identity (object-oriented programming) 、 Rhetorical question 、 Feature (linguistics)
摘要: Most text analysis and retrieval work to date has focused on the topic of a text; that is, what it is about. However, also contains much useful information in its style, or how written. This includes about author, purpose, feelings meant evoke, more. article develops new type lexical feature for use stylistic classification, based taxonomies various semantic functions certain choice words phrases. We demonstrate usefulness such features classification tasks determining author identity nationality, gender literary characters, text's sentiment (positivesnegative evaluation), rhetorical character scientific journal articles. further show functional aids gaining insight differences among different kinds texts. © 2007 Wiley Periodicals, Inc.