[目的/意義]隨著社會化媒體的興起,信息資源的數量呈現爆炸式增長,如何在海量的信息中幫助用戶發現有用的知識成為亟須解決的問題。互聯網上已經存在的各類用戶評論信息中蘊含著大量的可再開發的知識資源,包括用戶的個人信息、選擇偏好和消費習慣等,有助于解決"信息過載"問題。[方法/過程]文章通過對豆瓣電影評論信息進行細粒度的情感分析進而有效地獲取集體智慧,并且利用評論挖掘技術發掘用戶的偏好,為用戶選擇產品提供更加有效的推薦策略。[結果/結論]實驗表明,將大眾智慧與個性化服務兩者有機地結合起來,能夠真實地反映出不同用戶對于電影的感受特性,并為用戶觀影提供更加合理的參考。

[Purpose/significance] With the booming of social media,the number of information resources has been exploding. How to help users find useful information in vast amount of information becomes an urgent problem to be solved. The reviews of internet users contain lots of knowledge resources waited to be developed,including users' personal information,preferences,consumptive habits,and so on,which can solve the problem of information overload. [Method/process] The paper makes a finegrained sentimental analysis of film reviews on douban. com to get the collective wisdom effectively. Through mining technology from reviews to explore users' preference,the paper can provide a more effective recommendation strategy for users choosing products.[Result/conclusion]Experiment shows that the combination of the wisdom of the public and personalized service can truly reflect the feelings of different users and provide a more reasonable reference for the users.