AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
This paper proposes a novel approach to improving recommendation quality and value, especially in the environment of e-commerce, focusing on the following two issues

A hybrid online-product recommendation system: Combining implicit rating-based collaborative filtering and sequential pattern analysis

Electronic Commerce Research and Applications, no. 4 (2012): 309-317

被引用189|浏览7
EI

摘要

Many online shopping malls in which explicit rating information is not available still have difficulty in providing recommendation services using collaborative filtering (CF) techniques for their users. Applying temporal purchase patterns derived from sequential pattern analysis (SPA) for recommendation services also often makes users unh...更多

代码

数据

0
简介
  • In an age of information overload, the importance of personalized recommendation systems for online products and services is rapidly growing.
  • Such systems allow buyers to find what they want without wasting their time and enable sellers to provide buyers with the items they are likely to purchase, thereby furnishing benefits to both parties.
重点内容
  • In an age of information overload, the importance of personalized recommendation systems for online products and services is rapidly growing
  • This paper proposes a novel approach to improving recommendation quality and value, especially in the environment of e-commerce, focusing on the following two issues
  • We developed a recommendation system, called Hybrid Online-Product rEcommendation (HOPE), which integrates collaborative filtering (CF)-based recommendation using implicit rating and sequential pattern analysis (SPA)-based recommendation
  • In order to use the CF technique in such circumstance, this paper suggests a method of deriving implicit ratings of users on items from transaction data as an alternative to explicit ratings
  • Rather than suggesting a new approach to resolving these inherent problems we proposed a new CF-based approach which makes use of implicit rating information instead of explicit rating information that is required by the original CF technique, and we explained how implicit rating information can be computed from the transaction dataset
  • By using implicit ratings of users on items derived from transaction data, we could successfully apply CF technique to our data set, and by integrating implicit rating-based CF technique and SPA method, we were able to increase the accuracy of recommendation
  • We suggested an approach to obtaining better recommendation quality by integrating CF and SPA each of which considers the rating information of users on items and the associations among items, and proved it through our experiments
方法
  • The authors partitioned the data set into four parts, as shown in Fig. 3.
  • It was partitioned first by time into Part A and B, and second by random sampling of users into Parts C and D.
  • Part A consists of transaction data collected during the first 6 months and Part B during the second 6 months.
  • Part C consists of transaction data from 70% of the users, randomly chosen, and Part D the transaction data of the remaining users
结果
  • Experimental results and implications B

    the authors explain how the ideas proposed in Section 3 affect the accuracy of recommendations by analyzing the results from four experiments, and describe the implications of each idea.
结论
  • CF techniques have been used successfully to recommend various items such as movie and document
  • It requires that there should be many users who rated many items, so that items that are rated high by like-minded users of a user can be recommended to the user.
  • It has inherent problems such as the new user problem, the new item problem and the sparsity problem.
  • The authors suggested an approach to obtaining better recommendation quality by integrating CF and SPA each of which considers the rating information of users on items and the associations among items, and proved it through the experiments
总结
  • Introduction:

    In an age of information overload, the importance of personalized recommendation systems for online products and services is rapidly growing.
  • Such systems allow buyers to find what they want without wasting their time and enable sellers to provide buyers with the items they are likely to purchase, thereby furnishing benefits to both parties.
  • Objectives:

    Applying temporal purchase patterns derived from sequential pattern analysis (SPA) for recommendation services often makes users unhappy with the inaccurate and biased results obtained by not considering individual preferences.
  • Methods:

    The authors partitioned the data set into four parts, as shown in Fig. 3.
  • It was partitioned first by time into Part A and B, and second by random sampling of users into Parts C and D.
  • Part A consists of transaction data collected during the first 6 months and Part B during the second 6 months.
  • Part C consists of transaction data from 70% of the users, randomly chosen, and Part D the transaction data of the remaining users
  • Results:

    Experimental results and implications B

    the authors explain how the ideas proposed in Section 3 affect the accuracy of recommendations by analyzing the results from four experiments, and describe the implications of each idea.
  • Conclusion:

    CF techniques have been used successfully to recommend various items such as movie and document
  • It requires that there should be many users who rated many items, so that items that are rated high by like-minded users of a user can be recommended to the user.
  • It has inherent problems such as the new user problem, the new item problem and the sparsity problem.
  • The authors suggested an approach to obtaining better recommendation quality by integrating CF and SPA each of which considers the rating information of users on items and the associations among items, and proved it through the experiments
表格
  • Table1: An example of implicit rating data derived from an original transaction data
  • Table2: The integration of results from CF-based and SPA-based recommendation system
Download tables as Excel
基金
  • This research is supported by 2011 academic research funding from the Hanbat National University
研究对象与分析
users: 1000
The entire data set was collected from August 16, 2008 through August 15, 2009 (12 months) in four tables, for customers, sellers, products, and purchases. Since most recommendation systems have difficulty in recommending items to users who are involved in a small number of transactions, we focused on the users who have purchased more than 30 times among total 1000 users. As a result, 16,486 transactions of 247 users on 1911 items were used in our experiments.

Before conducting our experiments, we partitioned our data set into four parts, as shown in Fig. 3

users: 247
Since most recommendation systems have difficulty in recommending items to users who are involved in a small number of transactions, we focused on the users who have purchased more than 30 times among total 1000 users. As a result, 16,486 transactions of 247 users on 1911 items were used in our experiments.

Before conducting our experiments, we partitioned our data set into four parts, as shown in Fig. 3
. It was partitioned first by time into Part A and B, and second by random sampling of users into Parts C and D

users: 1000
The entire data set was collected from August 16, 2008 through August 15, 2009 (12 months) in four tables, for customers, sellers, products, and purchases. Since most recommendation systems have difficulty in recommending items to users who are involved in a small number of transactions, we focused on the users who have purchased more than 30 times among total 1000 users. As a result, 16,486 transactions of 247 users on 1911 items were used in our experiments

users: 247
Since most recommendation systems have difficulty in recommending items to users who are involved in a small number of transactions, we focused on the users who have purchased more than 30 times among total 1000 users. As a result, 16,486 transactions of 247 users on 1911 items were used in our experiments. Before conducting our experiments, we partitioned our data set into four parts, as shown in Fig. 3

引用论文
  • Adomavicius, G., and Kwon, Y. New recommendation techniques for multicriteria rating systems. IEEE Intelligent Systems, 22, 3, 2007, 48–55.
    Google ScholarLocate open access versionFindings
  • Adomavicius, G., and Tuzhilin, A. Towards the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering, 17, 6, 2005, 734–749.
    Google ScholarLocate open access versionFindings
  • Aggarwal, C. C., Procopiuc, C., and Yu, P. S. Finding localized associations in market basket data. IEEE Transactions on Knowledge and Data Engineering, 14, 1, 2002, 51–62.
    Google ScholarLocate open access versionFindings
  • Albadvi, A., and Shahbazi, M. A hybrid recommendation technique based on product category attributes. Expert Systems with Applications, 36, 9, 2009, 11480–11488.
    Google ScholarFindings
  • Baeza-Yates, R. A., and Ribeiro-Neto, B. Modern Information Retrieval. AddisonWesley Longman Publishing Co., Inc., Boston, MA, USA, 1999.
    Google ScholarFindings
  • Balabanovic, M., and Shoham, Y. Content-based, collaborative recommendation. Communications of the ACM, 40, 3, 1998, 66–72.
    Google ScholarLocate open access versionFindings
  • Belkin, N. J., and Croft, W. B. Information filtering and information retrieval: two sides of the same coin. Communications of the ACM, 35, 12, 1992, 29–38.
    Google ScholarLocate open access versionFindings
  • Billsus, D., and Pazzani, M. J. Learning collaborative information filters. In Proceedings of the Fifteenth International Conference on Machine Learning, 1998.
    Google ScholarLocate open access versionFindings
  • Billsus, D., Brunk, C. A., Evans, C., Gladish, B., and Pazzani, M. Adatpvie interfaces for ubiquitous web access. Communications of the ACM, 45, 5, 2002, 34–38.
    Google ScholarLocate open access versionFindings
  • Chen, L. S., Hsu, F. H., Chen, M. C., and Hsu, Y. C. Developing recommender systems with the consideration of product profitability for sellers. Information Sciences, 178, 4, 2008, 1032–1048.
    Google ScholarLocate open access versionFindings
  • Cheung, K. W., Kwok, J. T., Law, M. H., and Tsui, K. C. Mining customer product ratings for personalized marketing. Decision Support Systems, 35, 2, 2003, 231– 243.
    Google ScholarLocate open access versionFindings
  • Cho, Y. H., Kim, J. K., and Kim, S. H. A personalized recommender system based on web usage mining and decision tree induction. Expert systems with Applications, 23, 3, 2002, 329–342.
    Google ScholarLocate open access versionFindings
  • Getoor, L., and Sahami, M. Using probabilistic relational models for collaborative filtering. In Workshop on Web Usage Analysis and User Profiling (WEBKDD’99), 1999.
    Google ScholarFindings
  • Goldberg, D., Nichols, D., Oki, B. M., and Terry, D. Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35, 12, 1992, 61–70.
    Google ScholarLocate open access versionFindings
  • Goldberg, K., Roeder, T., Gupta, D., and Perkins, C. Eigentaste: a constant time collaborative filtering algorithm. Information Retrieval, 4, 2, 2001, 133–151.
    Google ScholarLocate open access versionFindings
  • Hofmann, T. Collaborative filtering via gaussian probabilistic latent semantic analysis. In Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2003, 259–266.
    Google ScholarLocate open access versionFindings
  • Hofmann, T. Latent semantic models for collaborative filtering. ACM Transactions on Information Systems, 22, 1, 2004, 89–115.
    Google ScholarLocate open access versionFindings
  • Huang, C. L., and Huang, W. L. Handling sequential pattern decay: developing a twostage collaborative recommendation system. Electronic Commerce Research and Applications, 8, 3, 2009, 117–129.
    Google ScholarLocate open access versionFindings
  • Jeong, B., Lee, J., and Cho, H. An iterative semi-explicit rating method for building collaborative recommender systems. Expert Systems with Applications, 36, 3, 2009, 6181–6186.
    Google ScholarLocate open access versionFindings
  • Jiang, Y., Shang, J., and Liu, Y. Maximizing customer satisfaction through an online recommendation system: a novel associative classification model. Decision Support Systems, 48, 3, 2009, 470–479.
    Google ScholarLocate open access versionFindings
  • Joaquin, D., and Naohiro, I. Memory-based weighted-majority prediction. In ACM SIGIR’ 99 Workshop on Recommender Systems: Algorithms and, Evaluation, 1999.
    Google ScholarLocate open access versionFindings
  • Kim, H. N., Ji, A. T., Ha, I., and Jo, G. S. Collaborative filtering based on collaborative tagging for enhancing the quality of recommendation. Electronic Commerce Research and Applications, 9, 1, 2010, 73–83.
    Google ScholarLocate open access versionFindings
  • Kim, H. K., Kim, J. K., and Ryu, Y. U. Personalized recommendation over a customer network for ubiquitous shopping. IEEE Transactions on Services Computing, 2, 2, 2009, 140–151.
    Google ScholarLocate open access versionFindings
  • Kim, J. K., Kim, H. K., and Cho, Y. H. A user-oriented contents recommendation system in peer-to-peer architecture. Expert Systems with Applications, 34, 1, 2008, 300–312.
    Google ScholarLocate open access versionFindings
  • Kumar, R., Raghavan, P., Rajagopalan, S., and Tomkins, A. Recommender systems: a probabilistic analysis. Journal of Computer and System Science, 63, 1, 2001, 42– 61.
    Google ScholarLocate open access versionFindings
  • Kwon, K., Cho, J., and Park, Y. Multidimensional credibility model for neighbor selection in collaborative recommendation. Expert Systems with Applications, 36, 3, 2009, 7114–7122.
    Google ScholarLocate open access versionFindings
  • Lang, K. NewsWeeder: learning to filter netnews. In Proceedings of the Twelfth International Conference on, Machine Learning, 1995.
    Google ScholarLocate open access versionFindings
  • Lee, J. S., and Olafsson, S. Two-way cooperative prediction for collaborative filtering recommendations. Expert Systems with Applications, 36, 3, 2009, 5353–5361.
    Google ScholarLocate open access versionFindings
  • Lee, K. C., and Kwon, S. Online shopping recommendation mechanism and its influence on consumer decisions and behavior: a causal map approach. Expert Systems with Applications, 35, 4, 2008, 1567–1574.
    Google ScholarLocate open access versionFindings
  • Lee, S. K., Cho, Y. H., and Kim, S. H. Collaborative filtering with ordinal scale-based implicit ratings for mobile music recommendations. Information Sciences, 180, 11, 2010, 2142–2155.
    Google ScholarLocate open access versionFindings
  • Lee, T. Q., Park, Y., and Park, Y. T. A time-based approach to effective recommender systems using implicit feedback. Expert Systems with Applications, 34, 4, 2008, 3055–3062.
    Google ScholarLocate open access versionFindings
  • Liang, T. P., Yang, Y. F., Chen, D. N., and Ku, Y. C. A semantic-expansion approach to personalized knowledge recommendation. Decision Support Systems, 45, 3, 2008, 401–412.
    Google ScholarLocate open access versionFindings
  • Liu, D. R., Lai, C. H., and Lee, W. J. A hybrid of sequential rules and collaborative filtering for product recommendation. Information Sciences, 179, 20, 2009, 3505–3519.
    Google ScholarLocate open access versionFindings
  • Liu, Z., Qu, W., Li, H., and Xie, C. A hybrid collaborative filtering recommendation mechanism for P2P networks. Future Generation Computer Systems, 26, 8, 2010, 1409–1417.
    Google ScholarLocate open access versionFindings
  • Marlin, B. Modeling user rating profiles for collaborative filtering. Advances in Neural Information Processing Systems, 16, 2003, 627–634.
    Google ScholarLocate open access versionFindings
  • Mooney, R. J., and Roy, L. Content-based book recommending using learning for text categorization. In ACM SIGIR ‘99 Workshop on Recommender Systems: Algorithms and Evaluation, Berkeley, CA, 1999.
    Google ScholarLocate open access versionFindings
  • Nakamura, A., and Abe, N. Collaborative filtering using weighted majority prediction algorithm. In 15th International Conference of Machine Learning (ICML ‘98), Madison, Wisconsin, USA, 1998, 395–403.
    Google ScholarLocate open access versionFindings
  • Park, Y. J., and Chang, K. N. Individual and group behavior-based customer profile model for personalized product recommendation. Expert Systems with Applications, 36, 2, 2009, 1932–1939.
    Google ScholarFindings
  • Pavlov, D. Y., and Pennock, D. M. A maximum entropy approach to collaborative filtering in dynamic, sparse, high-dimensional domains. Advances in Neural Information Processing Systems, 15, 2002, 1441–1448.
    Google ScholarLocate open access versionFindings
  • Pazzani, M., and Billsus, D. Learning and revising user profile: the identification of interesting web sites. Machine Learning, 27, 3, 1997, 313–331.
    Google ScholarLocate open access versionFindings
  • Pennock, D. M., Horvitz, E., Lawrence, S., and Giles, C. L. Collaborative filtering by personality diagnosis: a hybrid memory- and model-based approach. In Proceedings of the 16th Conference on Uncertainty in, Artificial Intelligence, 1999, 473–480.
    Google ScholarLocate open access versionFindings
  • Resnick, P., Iacovou, N., Suchak, M., Bergstrom, P., and Riedl, J. GroupLens: An open architecture for collaborative filtering of netnews. In Proceedings of the 1994 ACM Conference on Computer Supported Cooperative Work, 1994, 175–186.
    Google ScholarLocate open access versionFindings
  • Ricci, F., and Nguyen, Q. N. Acquiring an revising preferences in a critique-based mobile recommender system. IEEE Intelligent Systems, 22, 3, 2007, 22–29.
    Google ScholarLocate open access versionFindings
  • Russell, S., and Yoon, V. Applications of wavelet data reduction in a recommender system. Expert Systems with Applications, 34, 4, 2008, 2316–2325.
    Google ScholarLocate open access versionFindings
  • Salter, J., and Antonopoulos, N. Cinema screen recommender agent: combining collaborative and content-based filtering. IEEE Intelligent Systems, 21, 1, 2006, 35–41.
    Google ScholarLocate open access versionFindings
  • Salton, G. Automatic Text Processing. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 1988.
    Google ScholarFindings
  • Shani, G., Heckerman, D., and Brafman, R. I. An MDP-based recommender system. Journal of Machine Learning Research, 6, 2, 2002, 1265–1295.
    Google ScholarLocate open access versionFindings
  • Shardanand, U., and Maes, P. Social information filtering algorithms for automating ‘‘word of mouth’’. In Proceedings of the SIGCHI Conference on Human Factors in, Computing Systems, 1995, 210–217.
    Google ScholarLocate open access versionFindings
  • Shih, Y. Y., and Liu, D. R. Product recommendation approaches: collaborative filtering via customer lifetime value and customer demands. Expert Systems with Applications, 35, 1–2, 2008, 350–360.
    Google ScholarLocate open access versionFindings
  • Si, L., and Jin, R. Flexible mixture model for collaborative filtering. In Proceedings of 20th International Conference on, Machine Learning, 2003.
    Google ScholarLocate open access versionFindings
  • Su, J. H., Wang, B. W., Hsiao, C. Y., and Tseng, V. S. Personalized rough-set-based recommendation by integrating multiple contents and collaborative information. Information Sciences, 180, 1, 2010, 113–131.
    Google ScholarLocate open access versionFindings
  • Symeonidis, P., Nanopoulos, A., and Manolopoulos, Y. Providing justifications in recommender systems. IEEE Transactions on Systems, Man and Cybernetics – Part A: Systems an Humans, 38, 6, 2008, 1262–1272.
    Google ScholarLocate open access versionFindings
  • Wang, Y., Dai, W., and Yuan, Y. Website browsing aid: a navigation graph-based recommendation system. Decision Support Systems, 45, 3, 2008, 387–400.
    Google ScholarLocate open access versionFindings
  • Wei, C. P., Yang, C. S., and Hsiao, H. W. A collaborative filtering-based approach to personalized document clustering. Decision Support Systems, 45, 3, 2008, 413– 428.
    Google ScholarLocate open access versionFindings
  • Yu, K., Schwaighofer, A., Tresp, V., Xu, X., and Kriegel, H. P. Probabilistic memorybased collaborative filtering. IEEE Transactions on Knowledge and Data Engineering, 16, 1, 2004, 56–69.
    Google ScholarLocate open access versionFindings
  • Yu, K., Xu, X., Tao, J., Ester, M., and Kriegel, H. P. Instance selection techniques for memory-based collaborative filtering. In Proceeding of the Second SIAM International Conference on Data Mining (SDM’ 02), 2002.
    Google ScholarLocate open access versionFindings
  • Zenebe, A., and Norcio, A. F. Representation, similarity measures and aggregation methods using fuzzy sets for content-based recommender systems. Fuzzy Sets and Systems, 160, 1, 2009, 76–94.
    Google ScholarLocate open access versionFindings
  • Zhen, L., Jiang, Z., and Song, H. Distributed recommender for peer-to-peer knowledge sharing. Information Sciences, 180, 18, 2010, 3546–3561.
    Google ScholarLocate open access versionFindings
您的评分 :
0

 

标签
评论
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科