基于多模型的移动电子商务推荐系统设计与实现
|
摘要
随着Internet的普及和应用，电子商务因为其成本低廉、便捷、快速、不受时间和空间的限制等优点已在全球流行。电子商务在为用户提供更多选择的同时，其结构也日益更加复杂。一方面，用户面对大量的商品信息，很难快速找到自己真正需要的商品；另一方面，商家也无法与消费者面对面的交流。个性化的电子商务推荐系统能根据用户行为特征为用户提供一对一的服务，快速帮助用户找到所需的商品，从而顺利完成购物过程。商家通过推荐系统能提高电子商务系统销售能力，保持与客户的联系，提高用户忠诚度和满意度。
本文通过对当前B2C网站的电子商务个性化推荐系统分析，提出一种B2C模式下的多模型推荐系统（MMRS）的设计及实现，该系统通过对用户购物历史记录、Wap元数据以及用户注册信息处理，运用关联、聚类的方法，最后给出商品的推荐结果。这种对不同用户的多模型的推荐方案，即使新老用户由于信息的不同，都能够产生有效的推荐，并能够对新产品产生推荐。文中在推荐算法上做了一定改良，最后利用同组同学的Wap电子商务网站测试数据，对MMRS系统进行验证，发现改良后的算法能收到比较好的效果。
With the fast development of Internet，E-commerce has become more and morepopular all over the world. Business hence could overcome spatial and temporal barriersand are now capable of serving customers electronically and intelligently. However, theexponentially increasing amount of data and information along with the rapid expansion ofbusiness web sites and information systems make business hard to manage. On the otherhand，it is also difficult for customers to find the products they want. For these reasons，thepersonalized recommendation system arises at the right moment, which providescustomers one-to-one service based on their past behavior and reference from other userswith similar preferences. Many companies nowadays are using this system to retainexisting customers and attract new ones.
The article proposes a new multi-models recommendation system's (MMRS) designand realization based on the current B2C website electronic commerce personalizationrecommendation system's analysis. The system deals with purchasing history, Wap dataand user's registration information, uses the associate rule and cluster method,recommending the result in the way of commodities. The plan can give effectiverecommendation and new kinds of commodities to new and old users according to thedifferent information. There is some improvement in recommending calculate and testing itwith some data from a Wap site by my partner, and carrying on a verification to the MMRSsystem, and at the end we can find the improved calculate can receive good effect.
本文通过对当前B2C网站的电子商务个性化推荐系统分析，提出一种B2C模式下的多模型推荐系统（MMRS）的设计及实现，该系统通过对用户购物历史记录、Wap元数据以及用户注册信息处理，运用关联、聚类的方法，最后给出商品的推荐结果。这种对不同用户的多模型的推荐方案，即使新老用户由于信息的不同，都能够产生有效的推荐，并能够对新产品产生推荐。文中在推荐算法上做了一定改良，最后利用同组同学的Wap电子商务网站测试数据，对MMRS系统进行验证，发现改良后的算法能收到比较好的效果。
With the fast development of Internet，E-commerce has become more and morepopular all over the world. Business hence could overcome spatial and temporal barriersand are now capable of serving customers electronically and intelligently. However, theexponentially increasing amount of data and information along with the rapid expansion ofbusiness web sites and information systems make business hard to manage. On the otherhand，it is also difficult for customers to find the products they want. For these reasons，thepersonalized recommendation system arises at the right moment, which providescustomers one-to-one service based on their past behavior and reference from other userswith similar preferences. Many companies nowadays are using this system to retainexisting customers and attract new ones.
The article proposes a new multi-models recommendation system's (MMRS) designand realization based on the current B2C website electronic commerce personalizationrecommendation system's analysis. The system deals with purchasing history, Wap dataand user's registration information, uses the associate rule and cluster method,recommending the result in the way of commodities. The plan can give effectiverecommendation and new kinds of commodities to new and old users according to thedifferent information. There is some improvement in recommending calculate and testing itwith some data from a Wap site by my partner, and carrying on a verification to the MMRSsystem, and at the end we can find the improved calculate can receive good effect.
引文
[1] Schafer,J.B., Konstan,J.A., Riedl.J.Recommender Systems in E-Commerce. In ACMConference on Electronic commerce (EC99),1999.
[2] Schafer,J.B., Konstan,J.A.,Riedl.J.E-Commerce Recommendation Applications. DataMining and Knowledge Discovery,2001.
[3] Herlocker,J.Understanding and Improving Automated Collaborative Filtering Systems.PH.D. Thesis, Computer Science DePt., University of Minnesota,2000.
[4] Kohvai,R..Mining E-commerce Data the Good, the Bad, and the Ugly. Invitedpaper/talk at KDD2001industrial track,2001.
[5] S.Guha,R.Rastogi.,K.shim.Cure: An efficient clustering algorithm for largedatabases.Proc.1998ACM_SIMOD Int. Conf. Management of Data,1998,6:73-84.
[6]刘勘等.一种基于排序子空间的高维聚类算法及其可视化研究.计算机研究与发展,2003,40(10):1509-1513.
[7]付淇,李正凡.基于CLIQUE的聚类算法研究.华东交通大学学报,2006,23(5):79-82.
[8]黄光球,靳峰,彭绪友.基于兴趣度的协同过滤商品推荐系统模型.微电子学与计算机,2005,22(3):5-8.
[9] R.Agrawal, R.Srikant.Fast algorithms for mining association rules in large database.InProceedings of the20th VLDB Conference,Santiago,1994,12(8):22-30.
[10] Michael.O, Neil.H, Nicholas.K, et.al.Collaborative recommendation a robustnessanalysis. ACM Transactions on Internet Technology,Special Issue of MachineLearning for the Internet,2002,16(10):77-85.
[11] Gordon.S, Linoff.Michael, J.A.Berry.Web数据挖掘.沈钧毅,宋擒豹,燕彩蓉等译.北京:电子工业出版社,2004,25-34,99-104.
[12]（美）MICHAEL J.A.BERRY, GORDONS.LINOFF.数据挖掘技术在市场营销、销售与客户关系管理领域的应用,北京-机械工业出版社,2006.
[13]刘旭东.个性化网页推荐系统在电子商务中的设计与应用.烟台职业学院学报,2008,14(4)
[2] Schafer,J.B., Konstan,J.A.,Riedl.J.E-Commerce Recommendation Applications. DataMining and Knowledge Discovery,2001.
[3] Herlocker,J.Understanding and Improving Automated Collaborative Filtering Systems.PH.D. Thesis, Computer Science DePt., University of Minnesota,2000.
[4] Kohvai,R..Mining E-commerce Data the Good, the Bad, and the Ugly. Invitedpaper/talk at KDD2001industrial track,2001.
[5] S.Guha,R.Rastogi.,K.shim.Cure: An efficient clustering algorithm for largedatabases.Proc.1998ACM_SIMOD Int. Conf. Management of Data,1998,6:73-84.
[6]刘勘等.一种基于排序子空间的高维聚类算法及其可视化研究.计算机研究与发展,2003,40(10):1509-1513.
[7]付淇,李正凡.基于CLIQUE的聚类算法研究.华东交通大学学报,2006,23(5):79-82.
[8]黄光球,靳峰,彭绪友.基于兴趣度的协同过滤商品推荐系统模型.微电子学与计算机,2005,22(3):5-8.
[9] R.Agrawal, R.Srikant.Fast algorithms for mining association rules in large database.InProceedings of the20th VLDB Conference,Santiago,1994,12(8):22-30.
[10] Michael.O, Neil.H, Nicholas.K, et.al.Collaborative recommendation a robustnessanalysis. ACM Transactions on Internet Technology,Special Issue of MachineLearning for the Internet,2002,16(10):77-85.
[11] Gordon.S, Linoff.Michael, J.A.Berry.Web数据挖掘.沈钧毅,宋擒豹,燕彩蓉等译.北京:电子工业出版社,2004,25-34,99-104.
[12]（美）MICHAEL J.A.BERRY, GORDONS.LINOFF.数据挖掘技术在市场营销、销售与客户关系管理领域的应用,北京-机械工业出版社,2006.
[13]刘旭东.个性化网页推荐系统在电子商务中的设计与应用.烟台职业学院学报,2008,14(4)