國立臺灣大學資訊管理學系Tsay, Yih-KuenYih-KuenTsayChen, Yu-FangYu-FangChen2006-09-272018-06-292006-09-272018-06-29http://ntur.lib.ntu.edu.tw//handle/246246/20060927122917570645We propose and explore a novel approach,called the se- quence model,to text retrieval.The model di ffers from classical ones in the extent of how positional information of term occurrences is used for relevance judgment.In the sequence model,documents and queries are viewed as sequences of term-position pairs and the relevance of a doc- ument to a query is judged by the similarity between their respective rep- resentative sequences.We suggest three primitive measures of sequence similarity,each capturing a distinct aspect of resemblance between two sequences.These similarity measures can be combined in various ways to suit di fferent information needs.We have developed a prototype system with the sequence model as its core.Experimental results show that our sequence-based approach is often more e ffective than appearance-based approaches.application/pdf319570 bytesapplication/pdfzh-TWotherhttp://ntur.lib.ntu.edu.tw/bitstream/246246/20060927122917570645/1/00000000000000000000002.pdf