隨著部落格愈來愈受歡迎，使用者撰寫產品評論的數量也大量地被發表，使得部落格成為最豐富的產品評論意見來源。但一般人在撰寫部落格文章時，比較偏向口語和日記的形式，用字遣詞因人而異，文章主題也較發散。然而這樣的形式，使得其他新的消費者不易很快地從文章中取得他們所想要的資訊，業者也同樣地不容易追蹤客戶的意見。因此，如何去整合和分析部落格上大量使用者的產品評論並協助消費者獲得有關產品意見的結論已經是一個重要的議題。
過去在意見情緒分析上，有許多文獻將研究重點放在英文評論上。對於擷取產品特色與意見，辨別意見的語義傾向，以及產品特色與意見的結合有著豐富的研究成果。然而，同樣的方法卻不一定適用於中文評論上，所得到的結果也不一定能像在英文評論上有著優秀的表現。我們的目標是發展一個適用於中文評論的意見分析系統，此系統可以從部落格使用者評論中，擷取使用者關心的產品特色，並根據意見總是伴隨著產品特色出現的相依關係，針對此特色偵測使用者所表達的意見，且分析意見的語義傾向。經過整合後，呈現結果給使用者。根據實驗結果，證實我們所發展的系統能夠達到有用且顯著的效能改進。

As Weblog is becoming more and more popular, a large number of product reviews written by customers have been published, and this makes Weblog to become the largest resource of product reviews. However, diverse personal writing styles and fuzzy topics make it be time-consuming for consumers and sellers to read these reviews. Therefore, how to analyze and integrate the product reviews in Weblog and provide help for consumers to obtain inferences has become an important issue.
In the past, there are many documents focused on English reviews for sentiment analysis. There contains abundant research results with extracting features and opinions, identifying semantic orientation, and associating features with opinions. Even if it has excellent performance on English reviews, the approach is not as better on Chinese reviews. In this paper, our target is developing a sentiment analysis system that is suitable for Chinese reviews. This system would extract features that users are interested in from the reviews and detect the opinions with semantic orientations according to the dependency in features and opinions. Finally, we present the integrated results to users. Our experiments show that the derived system can effectively measure the dependency between features and opinions. The prominent performance on review sentiment analysis also validates the applicability of the proposed method.

[1] P. Chaovalit and L. Zhou. Movie Review Mining: A Comparison between Supervised and Unsupervised Classification Approaches. In proceedings of HICSS 2005.
[2] H. Cui, V. Mittal, and M. Datar. Comparative Experiment on Sentiment Classification for Online Product Reviews. In proceedings of AAAI 2006
[3] R. Fagin, R. Kumar, and D. Sivakumar. Comparing Top k List. In proceedings of SIAM 2003, pages 28-36.
[4] Z. Fei, J. Liu, and G. Wu. Sentiment Classification Using Phrase Patterns. In proceedings of CIT 2004, pages 1147-1152.
[5] N. Godbole, M. Srinivasaiah, S. Skiena. Large-Scale Sentiment Analysis for News and Blogs. In proceedings of ICWSM 2007, pages 219-222
[6] H. GUO, J. Jiang, G. Hu, and T. Zhang. Chinese named entity recognition based on multilevel linguistic features. In proceedings of IJCNLP 2004, pages 294-301.
[7] M. Hu and B. Liu. Mining and Summarizing Customer Reviews. In proceedings of KDD 2004.
[8] M. Hu and B. Liu. Mining Opinion Features in Customer Reviews. In proceedings of AAAI 2004, pages 754-760.
[9] J. Li and M. Sun. Experimental Study on Sentiment Classification of Chinese Review Using Machine Learning Techniques. In proceedings of NLP-KE 2007, pages 393-400.
[10] B. Liu, M. Hu, and J Cheng. Opinion Observer: Analyzing and Comparing Opinions on the Web. In proceedings of WWW 2005, pages 342-351.
[11] Y. Liu, X. Huang, A. An, and X. Yu. ARSA: A Sentiment-Aware Model for Predicting Sales Performance Using Blogs. In proceedings of SIGIR 2007, pages 607-614
[12] J. Read. Recognising Affect in Text using Pointwise-Mutual Information. Master Thesis, 2004
[13] B. Shi and K. Chang. Mining Chinese Reviews. In proceedings of ICDMW 2006.
[14] Q. Su, X. Xu, H. Guo, Z. Guo, X. Wu, X.Zhang, and B. Swen. Hidden Sentiment Association in Chinese Web Opinion Mining. In proceedings of WWW 2008, pages 959-968.
[15] P. D. Turney. Mining The Web for Synonyms: PMI-IR versus LSA on TOEFL. In proceedings of ECML 2001, pages 491-502.
[16] P. D. Turney. Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews. In proceeding of 40th Annual Meeting of the Association for Computational Linguistics (ACL 2002), pages 417-424.
[17] X. Wan. Using Bilingual Knowledge and Ensemble Techniques for Unsupervised Chinese Sentiment Analysis. In proceedings of EMNLP 2008, pages 553-561.
[18] B. Wang and H. Wang. Bootstrapping both Product Properties and Opinion Words from Chinese Reviews with Cross-Trainning. In proceedings of WI 2007
[19] C. Whitelaw, N. Garg, and S. Argamon. Using Appraisal Groups for Sentiment Analysis. In proceedings of CIKM 2005, pages 625-631.
[20] D. T. Wijaya and S. Bressa. A Random Walk on the Red Carpet: Rating Movies with User Reviews and PageRank. In proceedings of CIKM 2008, pages 951-960.
[21] S. Xu and M. Sun. Leveraging World Knowledge in Chinese Text Classification. In proceedings of ALPIT 2007, pages 33-38.
[22] J. Yao, G. Wu, J. Liu, and Y. Zheng. Using Bilingual Lexicon to Judge Sentiment Orientation of Chinese Words. In proceedings of CIT 2006.
[23] Q. Ye, B. Lin, and Y. J. Li. Sentiment Classification for Chinese Reviews: A Comparison Between SVM and Sentiment Approach. In proceedings of ICMLC 2005
[24] Q. Ye, Y. Li, and W. Shi. Sentiment Classification for Movie Reviews in Chinese by Improved Semantic Oriented Approach. In proceedings of HICSS 2006
[25] M. Zhang and X. Ye. A Generation Model to Unify Topic Relevance and Lexicon-based Sentiment for Opinion Retrieval. In proceedings of SIGIR 2008, pages 411-418.
[26] W. Zhang, C. Yu, and W. Meng. Opinion Retrieval from Blogs. In proceedings of CIKM 2007, pages 831-840.
[27] Y. Zheng, L. Ye, G. F. Wu, and X. Li. Extracting Product Features from Chinese Customer Reviews. In proceedings of ISKE 2008, pages 285-290.
[28] L. Zhuang, F. Jing, and A. Y. Zhu. Movie Review Mining and Summarization. In proceedings of CIKM 2006, pages 43-50.
[29] http://ckipsvr.iis.sinica.edu.tw/
[30] http://www.wretch.cc/blog/
[31] http://www.atmovies.com.tw/home/movie_homepage.html
[32] http://www.books.com.tw/books/index.php
[33] http://hk.music.yahoo.com/
[34] http://tw.movie.yahoo.com/
[35] http://www.kkbox.com.tw/funky/index.html