基于概率主题模型的游记文本知识挖掘-信号与信息处理专业论文.docxVIP

ABSTRACT With the pro叩erity oftravel and Web 2.0 technologies ，more and more people have willingness to share their travel experiences on the Web (e品， weblog s， forums， 也‘or Web 2.0 communities) in the form of t饵tual 阳velogues. These travelogues contain rich in如rmation， particularly including 1ωation-representative knowledge such as at衍actions，styles and activities. Such knowledge ωngr创tly facilitate other tourists trip planning ，if it can be correctly extracted and summarized. However 也‘ 电 since most 仕avelogues are noisy and lack of explicit destination recommendation or visual information， it is difficult for common users to utilize such knowledge effectively. In this pap町 to mine location守叩resentative knowledge from a large collection of travelogues ，we propose a probabilistic topic model，named as Location-Topic (LT) model. This model has the advantages of (1)differentiability between two kinds of topics，i.e.，local topics which characterize locations and global topics which ∞rrespondω ∞mmon themes co咱ωUITing with various locations in travelogues，and (2) r叩resen阳tion of locations in the local topic space to encode both location-representative knowledge and pairwise similarity between 10ω.tions. Some novel applications are developed based on the proposed model，including (1) destination recommendation und町 ei由时 oftwoαiteria ，i.e.，being similar ωa given destination or being relevant to a given travel intention， (2) destination summarization with r叩，resentative tags and relevant snippets，and (3) travelogue visualization with images depiωing the highlights in the text. 8ased on two large collections of travelogues，the proposed 仕amework is evaluated using b侃h objective and subjective approaches and shows promising results. 也 吗 KEY WORDS: Probabilistic ωpic model s， knowledge mining，travel planning， travelogues，us町-gen町ated ∞ntent 目录 第…章引宫 ‘1.1 研究背景……·川..川..川……… 川.川...…川..…..川...…川…..….. 川…..….川……川…川…..川.川…...….1 ‘ 1.2