1. You are Here：

Full metadata record
DC FieldValueLanguage
dc.contributor.author方策民en_US
dc.contributor.authorCe-Min Fangen_US
dc.contributor.author傅心家en_US
dc.contributor.authorHsin-Chia Fuen_US
dc.date.accessioned2014-12-12T02:27:38Z-
dc.date.available2014-12-12T02:27:38Z-
dc.date.issued2001en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#NT900392035en_US
dc.identifier.urihttp://hdl.handle.net/11536/68449-
dc.description.abstract本篇論文的主要目的，在於研究如何將文件處理技術應用在電視新聞內容分析上，精確而有效地針對以多媒體型式呈現的電視新聞做描述或索引， 方便使用者在已建構好的電視新聞檢索系統中做瀏覽的動作。我們使用bigram評分的方法，將電視新聞故事與文稿做配對的工作。當配對完成之後，更進一步針對文稿做關鍵詞的萃取，這個部份利用了詞庫比對和統計分析這兩種常見的方法做搭配。最後根據每則新聞故事所萃取出來的關鍵詞，建立每則新聞故事和其他新聞故事間的聯結，讓使用者能夠做主題式的瀏覽。zh_TW
dc.description.abstractThis thesis focuses on how to apply the text processing techniques to TV content analysis, then, we could precisely and efficiently describe or index TV news programs in multimedia form to facilitate user's browsing among TV news stories. We use bigram evaluation to match the TV news story and its corresponding text. When text is matched, lexicon matching and statistical analysis are used together to extract keyphrases from news text. Finally, according to the keyphrases extracted from each TV news story, we could construct the link among TV news stories, hence, user can browse various TV news story thematically.en_US
dc.language.isozh_TWen_US
dc.subject關鍵詞zh_TW
dc.subject相關新聞zh_TW
dc.subject文稿配對zh_TW
dc.subject電視新聞zh_TW
dc.subject多媒體zh_TW
dc.subject中文斷詞zh_TW
dc.subjectbigram matchingen_US
dc.subjectkeyword extractionen_US
dc.subjectrelated newsen_US
dc.subjectTV newsen_US
dc.subjectmultimediaen_US
dc.subjectChinese words segmentationen_US
dc.title電視新聞文稿之研究zh_TW
dc.titleThe Study of TV News Texten_US
dc.typeThesisen_US
dc.contributor.department資訊科學與工程研究所zh_TW
Appears in Collections:Thesis


Related Contents
Loading...