臺灣博碩士論文加值系統

在現今的電子商務環境之下，有著眾多的電子商務網站供使用者選擇，卻也降低了使用者對於網站之忠誠度。因此企業間為了因應商業競爭激烈，紛紛使用客戶關係管理策略，其中最常被使用的方法之一，就是「推薦系統(Recommendation system)」。利用推薦系統可減少客戶搜尋商品所耗之時間成本，並且可引導使用者對於企業特定欲銷售之商品的關注力，達成行銷策略中的目標市場行銷。此外，多數的大型電子商務網站已有多年的經營歷史，網站除了擁有交易資料，還有大量的網站瀏覽記錄，這些顧客的點擊順序，往往呈現使用者的消費意願。若能找出這些資料之間所隱含的消費傾向與習慣，將有助於網路行銷成效的提升。因此，本研究所提出的方法以點擊序列的時間先後順序來設定權重，並以滑動視窗(Sliding window)模型，將已有成交紀錄的顧客在不同時間區段內的點擊序列(Click sequence)作訓練分析，找出有效購買的點擊路徑。所提出的方法，透過實驗之呈現顧客瀏覽網頁的循序型別與購買交易決策的關聯性之綜整分析，掌握顧客動態的偏好特性，提升推薦效能。

The sheer volume of E-Commerce sites today provides vast amount of selections for customers but also reduces the loyalty of customers. Many enterprises have used customer relationship management to face the fierce competition and recommendation systems are widely used for customers’ retention. Recommendation systems may reduce the searching time cost of customers and direct users to the targeted products specified by the enterprises. That is, recommendation systems may increase the power of targeted marketing. In adition, a vast amount of operational data has been accumulated in e-commerce websites. These web-logs collect not only the committed transactions but also the clicking sequences, which indicates the consumers’ purchasing desire and thinking logic. Mining these data may reveal excellent target marketing strategies by exploring the relationships between a consumer’s tendency and his/her clicking path.
We propose an approach to adaptively adjust the weights of clicking sequences as time goes by. The approach uses a sliding window model, mines successful purchasing sequences from previous transactions, and discovers effective purchasing paths from sequences of different time periods. Experimental results report the relation of customer browsing sequences and purchasing transaction analysis, increase effectiveness of recommendations and marketing campaign by considering customer’s dynamic preferences.

目錄
摘要..........I
Abstract..........II
誌謝..........III
目錄 ..........V
圖目錄 ..........VII
表目錄 ..........IX
第一章 緒論 ..........1
1.1 研究背景 ..........2
1.2 研究動機 ..........3
1.3 研究目的 ..........5
1.4 研究範圍 ..........7
1.5 研究步驟 ..........9
1.6 論文架構 ..........11
第二章 文獻探討 ..........12
2.1 探勘應用 ..........12
2.2 個人化推薦系統 ..........14
2.3 資料探勘之序列樣式 ..........19
2.3.1 高頻序列樣式..........20
2.3.2 權重滑動視窗 (Weighted Sliding-Window)..........21
第三章 研究方法 ..........24
3.1 網頁點擊之序列探勘的交易推薦機制研究架構..........24
3.2 研究方法處理步驟..........25
3.2.1 資料前處理 (Purchase data pre-process)..........28
3.2.2 建構點擊序列購買決策樹(Click-sequence purchase decision tree, CSPD-tree)..........28
3.2.3 序列樣式之權重視窗支持度計算..........31
3.2.4 預測交易決策與推薦 ..........33
3.2.5 修剪點擊序列購買決策樹樹狀結構..........34
3.3 研究方法完整範例..........34
第四章 實驗結果 ..........38
4.1 實驗環境 ..........38
4.2 實驗資料 ..........38
4.3 實驗設計與分析 ..........40
第五章 結論..........52
5.1 研究貢獻..........52
5.2 未來研究工作..........53
參考文獻..........54
圖目錄
圖1.1研究範圍..........9
圖1.2 研究步驟..........11
圖2.1 界標視窗(Landmark window)..........22
圖2.2 滑動視窗(Sliding window)..........23
圖3.1 交易推薦機制架構..........25
圖3.2 有效購買路徑..........28
圖3.3 建構點擊序列購買決策樹..........30
圖3.4 處理CS<ABC{1,2}E{3}>..........30
圖3.5 處理CS<ACD{3}>..........30
圖3.6 處理CS<AEG{6}BC{2}>..........30
圖3.7 處理CS<ABG{1}E{2}>..........31
圖3.8 建構完成CSPD-tree圖 ..........35
圖3.9 目標行銷比對..........36
圖4.1 網站三層式架構..........39
圖4.2總序列數量對於執行時間之影響..........43
圖4.3 支持度對於點擊序列數量之影響..........44
圖4.4 最大序列長度對於點擊序列數量之影響..........45
圖4.5視窗尺寸對於點擊序列數量之影響..........46
圖4.6滑動視窗尺寸大小對於平均序列長度影響..........47
圖4.7有無權重值之點擊序列數量影響..........48
圖4.8權重值設定之影響..........50
表目錄
表3.1 顧客交易與點擊資料..........29
表3.2 購物資料前處理..........35
表3.3 CSPD-tree節點計算方式..........36
表4.1 實驗平台規格..........38
表4.2資料產生器之參數說明..........40
表4.2 實驗設計之內容..........41
表4.3網頁點擊之探勘方法組合..........51

1.吳肇銘、金志聿與林怡秀(2004)，「協同過濾技術在商品推薦系統上之應用與成效評估」，第十五屆ICIM國際資訊管理學術研討會論文集，台北，第1-12頁。
2.李御璽與顏秀珍(2008)，「網路交易型樣探勘技術之研究」，電子商務學報，第10卷，第4期，第989-1008頁。
3.柯常恩(2008)，以樹狀拆解為基礎探勘連續序列型樣之研究，碩士論文，南台科技大學資訊管理系，台南。
4.胡雅涵(2007)，使用以限制為基礎的序列規則方式的顧客購買行為研究，博士論文，國立中央大學資訊管理系，高雄。
5.張和逸(2008)，高效率探勘連續路徑序列型樣之研究，碩士論文，南台科技大學資訊管理系，台南。
6.莊清男(2005)，協同過濾式群體推薦，碩士論文，國立中央大學資訊管理系，高雄。
7.陳仕昇(2003)，序列樣式探勘之研究，碩士論文，國立中央大學資訊管理系，高雄。
8.陳正德(2004)，以項目為基礎的協同過濾應用於網路教材瀏覽推薦之研究，碩士論文，銘傳大學資訊工程系，台北。
9.陳正德(2004)，以項目為基礎的協同過濾應用於網路教材瀏覽推薦之研究，碩士論文。銘傳大學資訊工程系，台北。
10.黃柏文(2007)，漸進式網路交易型樣探勘技術之研究，碩士論文，銘傳大學資訊工程系，台北。
11.黃純敏、林重佑、黃進瑞（2013），「結合學習向量量化與協同過濾之交換混合式過濾電影推薦架構」，資訊管理學報，第二十卷，第四期，第423-448頁。
12.廖紹安(2011)，探勘封閉網路瀏覽型樣之研究，碩士論文，銘傳大學資訊工程系，台北。
13.賴雨廷(2001)，利用資料探勘技術應用於圖書館新書推薦之研究，碩士論文，國立中山大學資訊管理系，高雄。
14.謝百恩(2006)，有型樣限制的序列型樣探勘演算法，碩士論文，銘傳大學資訊工程系，台北。
15.羅健銘(1991)，協同過濾於網站推薦之研究，碩士論文，台北科技大學商業自動化與管理研究所，台北。
16.Agrawal, Rakesh, Faloutsos, Christos and Swami Arun(1993), “Efficient Similarity Search in Sequence Databases,” Proeedings of the 4th International Conference on Foundations of Data Organization and Algorithms, pp. 69-84, Chicago.
17.Agrawal, Rakesh, and Srikant, Ramakrishnan(1994), “Fast algorithms for mining association rules,” Proceedings of the 20th International Conference on Very Large Data Bases, pp. 487-499, Chile.
18.Agrawal, Rakesh, and Srikant, Ramakrishnan(1995), “Mining sequential patterns,” Proceedings of the 11th International Conference on Data Engineering, pp. 3-14, Taiwan.
19.Agrawal, Rakesh, and Srikant, Ramakrishnan(1996), “Mining sequential patterns: generalizations and performance improvements,” Proceedings of the 5th International Conference on Extending Database Technology, pp. 3-17, France.
20.Agrawal, Rakesh, Imieliński, Tomasz, and Swami, Arun(1993), "Mining association rules between sets of items in large databases," ACM SIGMOD Record, Vol. 22, No. 2, pp. 207-216.
21.Chang, Joong Hyuk(2011), "Mining weighted sequential patterns in a sequence database with a time-interval weight," Knowledge-Based Systems, Vol. 24, No. 1, pp. 1-9.
22.Chang, Joong Hyuk, and Lee, Won Suk(2006), “Finding recently frequent itemsets adaptively over online transactional data streams,” Information Systems, Vol. 31, No. 8, pp. 849-869.
23.Chen, Hui, Shu, LihChyun, Xia, Jiali and Deng, Qingshan(2012), “Mining frequent patterns in a varying-size sliding window of online transactional data streams,” Information Science, Vol. 215, No. 15, pp. 15-36.
24.Chen, Ming Syan, Park, Jong Soo, and Yu, Philip S.(1998), “Efficient data mining for path traversal patterns,” IEEE Transactions on Knowledge and Data Engineering, Vol. 10, No. 2, pp. 209-221.
25.Chen, Yen-Liang, and Huang, Tony Cheng-Kui (2005), "Discovering fuzzy time-interval sequential patterns in sequence databases," IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, pp. 959-972.
26.Choi, Keunho, Yoo, Donghee, Kim, Gunwoo and Suh, Yongmoo(2012), "A hybrid online-product recommendation system: Combining implicit rating-based collaborative filtering and sequential pattern analysis," Electronic Commerce Research and Applications, Vol. 11, No. 4, pp. 309-317.
27.Dai, Bi-Ru, and Jiang, Hung-Lin and Chung, Chih-Heng(2010), “Mining top-k sequential patterns in the data stream environment,” Proceedings of Technologies and Applications of Artificial Intelligence (TAAI), 2010 International Conference on, pp. 142-149, Taiwan.
28.Goksedef, Murat and Gunduz-Oguducu, Sule(2010), “Combination of web page recommender systems,” Expert Systems with Applications, Vol. 37, No. 4, pp. 2911-2922.
29.Han, Jiawei, Pei, jian, Mortazavi-Asl, Behzad, Chen, Qiming, Dayal, Umeshwar and Hsu, Meichun(2000), “FreeSpan: Frequent Pattern-Projected Sequential Pattern Mining,” Proceedings of ACM SIGKDD International Conference on Knowledge Discovery Data mining, pp. 355-359, USA.
30.Han, Jiawei, Dong, Guozhu and Yin, Yiwen(1999), “Efficient mining of partial periodic patterns in time series database,” Proeedings of 15th International Conference on Data Engineering, pp.106-115, Sydney.
31.Hill, William Colyer, Stead, Larry, Rosenstein, Mark, Furnas, George(1995), ” Recommending and evaluating choices in a virtual community of use,” Proceedings of ACM CHI’95 Conference on Human Factors in Computing Systems, pp. 194-201, New York.
32.Hoang, Thanh Lam, and Calders, Toon(2010), “Mining top-k frequent items in a data stream with flexible sliding windows,” Proceedings of 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 283-292, USA.
33.Jea, Kuen-Fang, and Li, Chao-Wei(2010), “A sliding-window based adaptive approximating method to discover recent frequent itemsets from data streams,” Lecture Notes in Engineering and Computer Science, Vol. 2180, No. 1, pp. 532-539.
34.Kim, Dohyun, and Yum, Bong Jin (2005), “Collaborative filtering based on iterative principal component analysis,” Expert Systems with Applications, Vol. 28, No. 4, pp. 823-830.
35.Kim, Yong Soo, and Bong-Jin Yum(2011), "Recommender system based on click stream data using association rule mining," Expert Systems with Applications, Vol. 38, No. 10, pp. 13320-13327.
36.Kim, Yong Soo, Yum, Bong Jin, Song, Junehwa and Kim, Su Myeon(2005), “Development of a recommender system based on navigational and behavioral patterns of customers in ecommerce sites,” Expert Systems with Applications, Vol. 28, No. 2, pp. 381-393.
37.Lang, Ken(1995), “NewsWeeder: Learning to filter netnews,” Proceedings of the 12th International Conference on Machine Learning, pp. 331-339, San Francisco.
38.Lawrence, Richard D., Almasi, George S., Kotlyar, Vladimir, Viveros, Marisa S. and Duri, Sastry(2001), “Personalization of Supermarket Product Recommendations,” Data Mining and Knowledge Discovery, Vol. 5, No. 1/2, pp. 11-32.
39.Li, Hua-Fu(2009), “A sliding window method for finding top-k path traversal patterns over streaming Web click-sequences,” Expert Systems with Applications, Vol. 36, No. 3, pp. 4382-4386.
40.Li, Hua-Fu(2009), “Mining top-k maximal reference sequences from streaming web click-sequences with a damped sliding window.” Expert Systems with Applications, Vol. 36, No. 8, pp. 11304-11311.
41.Li, Hua-Fu, and Lee, Suh-Yin(2008), “Approximate mining of maximal frequent itemsets in data streams with different window models,” Expert Systems with Applications, Vol. 35, No. 3, pp. 781-789.
42.Li, Hua-Fu, and Lee, Suh-Yin(2009), “Mining frequent itemsets over data streams using efficient window sliding techniques,” Expert Systems with Applications, Vol. 36, No. 2, pp. 1466-1477.
43.Li, Hua-Fu, Lee, Suh-Yin, and Shan, Man-Kwan(2005), “DSM-TKP: Mining top-k path traversal patterns over web click-streams,” Proceedings of the 2005 IEEE/WIC/ACM International Conference on Web Intelligence, pp. 326-329.
44.Mobasher, Bamshad, Jain, Namit, Han, Eui and Srivastava, Jaideep(1996), Web mining: pattern discovery from world wide web transactions, Technical Report, TR96-050, Department of Computer Science, University of Minnesota.
45.Pei, Jian, Han, Jiawei, Mortazavi-Asl, Behzad and Zhu, Hua(2000), “Mining access pattern efficiently from web logs,” Proceedings of Pacific-Asia Conference on Knowledge Discovery and Data Mining, pp. 396-407, Japan.
46.Pei, Jian, Han, Jiawei, Mortazavi-Asl, Behzad, Pinto, Helen, Chen, Qiming, Dayal, Umeshwar and Hsu, Meichun (2001), “PrefixSpan: mining sequential patterns efficiently by prefix-projected pattern growth,” Proceedings of the 17th International Conference on Data Engineering, pp. 215-224, Germany.
47.Pei, Jian, Han, Jiawei, Mortazavi-Asl, Behzad, Pinto, Helen, Chen, Qiming, Dayal, Umeshwar and Hsu, Meichun (2004), “Mining sequential patterns by pattern-growth: the prefixspan approach,” IEEE Transactions on Knowledge and Data Engineering, Vol. 16, No. 10, pp. 1424-1440.
48.Resnick, Paul (1997), “Recommender Systems,” Communications of the ACM, Vol. 40, No. 3, pp. 56-58.
49.Resnick, Paul, Iacovou, Neophytos, Suchak, Mitesh, Bergstrom, Peter and Riedl, John(1994), “GroupLens: An open architecture for collaborative filtering of netnews,” Proceedings of ACMCSCW’94 Conference on Computer-Supported Cooperative Work, pp. 175-186, USA.
50.Tanbeer, Syed Khairuzzaman, Ahmed, Chowdhury Farhan, Jeong, Byeong-Soo and Lee, Young-Koo(2009), “Sliding window-based frequent pattern mining over data streams,” Information Sciences, Vol. 179, No. 22, pp. 3843-3865.
51.Wong, Raymond Chi-Wing, and Fu, Ada Wai-Chee (2006), “Mining top-k frequent itemsets from data streams,” Data Mining and Knowledge Discovery, Vol. 13, No. 2, pp. 193-217.
52.Xiao, Yongqiao and Dunham, Margaret H.(2001), “Efficient mining of traversal patterns,” Data & Knowledge Engineering, Vol. 39, No. 2, pp. 191-214.
53.Yang, Kung-Jiuan, Hong, Tzung-Pei, Lan, Guo-Cheng and Chen, Yuh-Min(2014), “A two-phase approach for mining weighted partial periodic patterns,” Engineering Applications of Artificial Intelligence, Vol.30, pp.225-234.
54.Zaki, Mohammed Javeed (2001), “SPADE: an efficient algorithm for mining frequent sequences,” Machine Learning Journal, Vol. 42 No. 1/2, pp. 31-60.
55.Zhu, Yunyue, and Shasha, Dennis.(2002), "Statstream: Statistical monitoring of thousands of data streams in real time." Proceedings of the 28th international conference on Very Large Data Bases(VLDB), pp. 358-369, China.