跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 曾薇蓉
Wei-Jung Tseng
論文名稱： 刪除協同推薦系統中偏誤評估者之研究
Screening unfair raters from collaborative filtering recommendation system
指導教授： 許秉瑜
Ping-Yu Hsu
口試委員:
學位類別： 碩士
Master
系所名稱： 管理學院 - 企業管理學系
Department of Business Administration
畢業學年度： 94
語文別： 英文
論文頁數： 48
中文關鍵詞： 資料探勘 、協同過濾演算法 、推薦系統
外文關鍵詞： collaborative filtering, data mining, recommendation system
相關次數： 點閱：8  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 推薦系統目前廣泛的被使用在電子商務環境中，協同過濾演算法(Collaborative filtering algorithm)最常被使用，但協同過濾法有一些缺點，例如是新的使用者、新的推薦項目通常都沒辦法獲得準確推薦、資料庫稀疏的問題、以及惡意評比而影響推薦準確度，本研究改進協同過濾推薦的惡意評比問題，在方法上，提出在協同過濾演算法之前的過濾機制是以beta分配為基礎，此方法通常運用於線上拍賣系統中，本文藉此過濾機制以預測是否此機制可以有效的刪除惡意攻擊推薦系統的推薦準確度。實驗所使用的資料是以MovieLens的資料庫為主，在本實驗中把使用者按照電影種類來區分，個別觀察其推薦的項目、次序。
    實驗共分成兩個，第一個是混入人工製造的資料，此可視為惡意攻擊之評比；第二個是用原本的MovieLens的資料庫的資料，因此在實驗評估共分兩個，第一個是比較加入人工製造之資料是否會影響推薦的項目。第二是比較是否在兩者實驗刪除部分使用者後其推薦項目是否會被影響。本實驗結果是在實驗二，用MovieLens原始資料對刪除部分使用者前後的推薦項目並無太大差異，且在混入人工製造資料的實驗中過濾偽評比後可以完全刪除人工製造之評比外，其推薦項目和在使用原本的MovieLens的資料庫的評比尚未做刪除機制的結果亦無太大差異。

    Recommender systems are widely used in the e-commerce environment. Until recently, most used algorithms or collaborative filtering methods; but in these approaches, shortcomings: new use and item problem, as well as sparsity and attacks on recommender systems that will affect its precision. We propose a solution to solve problems caused by attacks on recommender systems using a beta reputation system in order to filter out unfair ratings in the recommender systems. This method is popular in the on-line marketplace but we have improved the ability to delete unfair rating existing in the recommender system though this filtering method. The dataset we used is MovieLens. We group users’ ratings by movie categories and observed the difference of recommendation items.
    There are two kinds of data content in our research; the first one is combines the MovieLens dataset with man-made ratings which can be seen as ratings by attacker. The second one is original MovieLens dataset. There are 2 evaluations: average difference in percentage of two experiments and average deleting users of two experiments.
    The advantage in our proposed method is that it filters out fake users in the recommender system. The results show that the accuracy and the performance are considerably improved.

    Abstract II Table of Contents III List of Figures V List of Tables VI Chapter1 Introduction 1 1-1 Background 1 1-2 Motivation 2 1-3 Thesis organization 3 Chapter2 Literature review 4 2-1 Recommender system 4 2-1-1 Collaborative filtering based Recommender systems 4 2-1-2 Memory-based (user-based) collaborative filtering approach 5 2-1-3 Model-based (item-based) collaborative filtering approach 9 2-1-4 Content-based recommender systems 11 2-2 Reputation systems 11 2-2-1 The Beta distribution 13 2-2-2 The beta reputation system 15 2-2-3 The reputation rating 16 2-2-4 Aggregating reputation rating 17 2-2-5 The reputation score 18 Chapter3 Research Methodology 20 3-1 Problem definition and description 20 3-2 Filtering Recommendation Algorithm 21 3-2-1 Converting ratings to binary rating matrix for each rater (user) 21 3-2-2 Filtering data with beta distribution 21 3-2-3 Inverting fair ratings into user-item matrix 25 Chapter4 Simulation results 26 4-1 Simulation datasets 26 4-2 Experimental procedure 26 4-3 Measuring metrics 28 4-4 Experimental results 29 Chapter5 Conclusions 34 References 35 Appendix 1 37 Appendix 2 42 Appendix 3 45

    A. Jøsang, S. H. a. E. F. (2003, May). Simulating the effect of reputation systems on e-markets. Paper presented at the In the proceedings of the First International Conference on Trust Management.
    A. Whitby, A. J. a. J. I. (2005). Filtering out unfair ratings in bayesian reputation systems. The Icfain Journal of Management Research, 4(2), 48-64.
    Andreas Schlosser, M. V., Lars Br uckner. (2005). Comparing and evaluating metrics for reputation systems by simulation. Paper presented at the A Workshop on Reputation in Agent Societies.
    B. Sarwar, G. K., J. Konstan, and J. Riedl. (2000). Application of dimensionality reduction in recommender systems: A case study. Paper presented at the in ACM WebKDD Workshop.
    Badrul Sarwar, G. K., Joseph Konstan, and John Riedl. (2001). Item-based collaborative filtering recommendation algorithms. Paper presented at the In Proceedings of the 10th International World Wide Web Conference, Hong Kong.
    Bhattacharjee, P. M. a. B. (2004). Using trust in recommender systems: An experimental analysis. Paper presented at the iTrust2004 International Conference.
    Chun Zeng, C.-X. X., Li-Zhu Zhou. (2003). Similarity measure and instance selection for collaborative filtering.
    Gediminas Adomavicius, a. A. T. (2005). Toward 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).
    Http://movielens.Umn.Edu.
    Ismail, A. J. a. R. (2002, 17-19 June). The beta reputation system. Paper presented at the In the proceedings of the 15th Bled Conference on Electronic Commerce, Bled, Slovenia.
    John S. Breese, D. H., Carl Kadie. (1998). Empirical analysis of predictive algorithms for collaborative filtering. Microsoft Research Microsoft Corporation One Microsoft Way Redmond.
    KARYPIS, M. D. a. G. (2004). Item-based top-n recommendation
    algorithms. ACM Transactions on Information Systems, 22(1), 143–177.
    Rong Jin, J. Y. C., Luo Si. (2004). An automatic weighting scheme for collaborative filtering. Paper presented at the SIGIR 2004.
    Yu Li, L. L., Li Xuefeng. (2005). A hybrid collaborative filtering method for multiple-interests and multiple-content recommendation in e-commerce. Expert Systems with Applications, 28, 67-77.

    QR CODE
    :::