随着互联网技术的广泛应用, 人们正陷入信息过载的困境。为解决信息过载问题, 人们采用推荐系统向用户提供建议, 帮助用户获得感兴趣的信息。在推荐系统中用户模型占有十分重要的地位, 其实现了用户的个性化表示。本文面向制造业领域的企业构建用户模型, 称其为企业模型, 采用文本挖掘技术获得与企业产品相关联的特征词, 采用主动学习技术辅助构建特征词之间的关联关系, 通过特征词及特征词之间的关联关系表示企业模型。构建的企业模型能够表示企业产品的特征, 从而表示企业在这些特征领域的信息需求, 为后续推荐系统的实现奠定了基础。
Due to the broad application of the internet, we are suffering from the problem of information overloading.To solve this problem, we recommend to users some suggestions through recommender system.User model is an important part of recommender system, which realizes the personalized expression for users.In this paper, by using the data mining method, the enterprise model is made to express the enterprises of manufacturing industry through the feature words and the relations among the feature words which are mined from the texts related to the enterprises′ products.The enterprise model is used to express the users′ needs and provides data foundation for the recommender system.
[1]余力, 刘鲁, 李雪峰.用户多兴趣下的个性化推荐算法研究[J].计算机集成制造系统, 2004, 10(12):1610-1616.
[2]Michael D.Ekstrand, Joseph A.Konstan, John T.Riedl.Collaborative filtering recommender systems[J].Foundations and Trends in Human-Computer Interaction, 2011, 4(2):81-173.
[3]Greg Linden, Brent Smith, Jeremy York.Recommendations item-to-item collaborative filtering[C].IEEE, 2003, 76-80.
[4]Manos Papagelis, Dimitris Plexousakis.Qualitative analysis of user-based and item-based prediction algorithms for recommendation agents[C].Engineering Applications of Artificial Intelligence, 2005, I8:781-789.
[5]Michael J.Pazzani1, Daniel Billsus.Content-based recommendation systems[C].Springer-Verlag Berlin Heidelberg, 2007:325-341.
[6]Marko Balabanovic, Yoav Shoham.Content-based collaborative filtering recommendation[C].Communications of the ACM, 1997, 40(3):66-72.
[7]Rakesh Agrawal, Tomasz Imielinski.Mining association rules between sets of items in large databases[C].ACM, 1993:1-10.
[8]Nick Craswell, Martin Szummer.Random walks on the click graph[C].SIGIR, 2007:1-6.
[9]吴丽花, 刘鲁.个性化推荐系统用户建模技术综述[J].情报学报, 2006, 25(1):55-66.
[10]David Vickrey, Oscar Kipersztok, Daphne Koller.An active learning approach to finding related terms[C].ACM, 2010:1-6.
[11]张晓莹, 张桂平, 王裴岩.领域本体构建中关系辅助判断技术研究[J].中国计算语言学研究前沿进展, 2011:276-282.
[12]孙浩.基于主动学习的文本过滤系统的研究与应用[D].北京:北京邮电大学, 2011:12-26.
[13]葛世伦.企业信息模型研究[J].华东船舶工业学报(自然科学版), 2001, 15(3):73-79.
[14]张桂平, 尹宝生.知识管理在军工企业中的应用[J].沈阳航空航天大学学报, 2010, 27(4):47-49.
[15]白宇, 王裴岩, 蔡东风.专利检索技术[J].沈阳航空航天大学学报, 2010, 27(4):50-53.
[16]刘建国, 周涛, 郭强, 等.个性化推荐系统评价方法综述[J].复杂系统与复杂性科学, 2009, 06(3):1-10.
