Full metadata record
DC FieldValueLanguage
dc.contributor.author盧俊錡en_US
dc.contributor.author李嘉晃en_US
dc.contributor.authorLee, Chia-Hoangen_US
dc.date.accessioned2014-12-12T01:34:44Z-
dc.date.available2014-12-12T01:34:44Z-
dc.date.issued2008en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT079657548en_US
dc.identifier.urihttp://hdl.handle.net/11536/43554-
dc.description.abstract隨著網際網路的蓬勃發展,部落格與討論版的興起,越來越多人在網路上發表自己對事物的意見以及看法。也因此,當購物者對某件商品下決策時,大部分人們在網路上對該產品的評價往往是很重要的參考依據。例如,一個網路的使用者在選擇要看甚麼電影之前,通常會先瀏覽熱門電影討論版參考看過該電影的使用者的評價來做為決定的因素。但對大部分使用者而言,要消化掉網路上大量對產品的評價資訊可能是相當耗時的。因此,在本篇論文中,我們透過自然語言處理以及資料探勘中的分群技術,來分析影評對該電影的評價是『好看』或『不好看』,並利用自動摘要技巧把影評中『好看』或『不好看』的句子擷取出來回饋給使用者。希望使用者透過我們的介面,可以在比較參考大量的評論資訊時,可以用更簡單,清楚,直覺的比較並做出決定。zh_TW
dc.description.abstractWith the rapid development of Internet and rise of BLOG and Discussion board , there are more and more people express their views or opinion on things on the internet . Thus , most of people’s appraisals on the web are significant information for customer making their decision . For example , people could Decided to go to the movies according to the existing appraisals on the web . But for most of user it is Time consumption to read all reviews on the movie Discussion board . In this Paper , we apply the Natural Language Processing (NLP) technology and classification technology to classify Text with two polarity : good or bad . Then we combine auto summarization technology to generalized a corresponding appraisal . We hope the user can make decision more rapid through the system which is designed by us .en_US
dc.language.isozh_TWen_US
dc.subject情感分類zh_TW
dc.subject情感探勘zh_TW
dc.subject自動摘要zh_TW
dc.subjectauto summarizationen_US
dc.subjectSentiment Classificationen_US
dc.title影評意見探勘及摘要之問答系統zh_TW
dc.titleQuestion Answering - Opinion mining And Auto summarization for Movie Reviewen_US
dc.typeThesisen_US
dc.contributor.department資訊科學與工程研究所zh_TW
Appears in Collections:Thesis


Files in This Item:

  1. 754801.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.