臺灣博碩士論文加值系統

使用資料探勘( Data Mining )之相關技術以探討電信產業訂單資料庫中客戶電信服務的特性是一項值得探討的研究，引用 ”Mining Frequent Patterns without Candidate Generation” 之理論，以 FP-tree 資料架構來構建客戶服務的資料，藉由 FP-Growth 演算法產生頻繁型樣，電信產業訂單資料的頻繁型樣是否存在，是否能夠藉此演算法之實作產生？產出的頻繁型樣是否有其價值？
同時藉由實作提出對此演算法的幾項改進，讓此演算法更能應用於所被探勘的電信產業訂單資料及提昇演算法的重要性及實用價值，並將所探勘出來之頻繁型樣可以彈性動態地予以調整，為滿足研究宗旨因此將客戶資料分類化、動態給定最低支持度 ，動態給定各分類的最低支持度等改善方法，期望藉由這些改善方法及程序能夠適當及有效地將 FP-tree及FP-Growth 演算法所產生過多的頻繁型樣適當修剪，並以 DGFP (Dynamically Generate Frequent Patterns )，一個互動模式動態產生頻繁型樣的工具，以方便使用者進行資料探勘。

Using data mining technique to explore the customer service characteristics from telecom business order processing database is an issue worth studying. This paper uses the “Mining Frequent Patterns without Candidate Generation” algorithm and uses FP-tree structure to construct customer frequent pattern data tree, and uses FP-growth method to generate the frequent patterns to see whether the frequent patterns can be discovered and are valuable. Also by making several improvements on the approaches, we are able to broaden and multiply the basic algorithm value, such as classifying the customer service data, applying constraints and different minimum support thresholds for each class to prune frequent patterns properly and effectively, and creating a DGFP ( Dynamically Generate Frequent Patterns ), a tool to let users interactively online generate frequent patterns dynamically.

第一章 緒論
1.1 前言………………………………………………………………………1
1.2 相關研究與回顧…………………………………………………………2
1.2.1資料探勘技術於電信產業之相關研究………………………………2
1.2.2資料探勘技術之相關研究……………………………………………2
1.3 研究動機…………………………………………………………………3
1.4 研究目的…………………………………………………………………3
1.5 內容簡介…………………………………………………………………5
第二章 基礎知識
2.1 資料探勘相關理論……………………………………………………5
2.1.1資料探勘的相關理論分類……………………………………………5
2.1.2.基本定義……………………………………………………………6
2.2 文獻探討………………………………………………………………7
2.3 被探勘的系統及其資料………………………………………………8
2.4 研究方法………………………………………………………………11
2.4.1.研究問題……………………………………………………………11
2.4.2.研究理論……………………………………………………………12
第三章 動態探勘頻繁型樣方法之實作
3.1 動態探勘頻繁型樣方法的實作程序…………………………………15
3.1.1第一期-探勘一般客戶群……………………………………………15
3.1.2第二期-探勘特定客戶群……………………………………………19
3.2 動態探勘頻繁型樣之方法理論………………………………………19
3.2.1第一期-探勘一般客戶群……………………………………………19
3.2.2第二期-探勘特定客戶群……………………………………………22
3.3開發環境與工具…………………………………………………………27
3.4系統程式運作流程………………………………………………………27
3.5系統功能及程式架構……………………………………………………28
3.5.1建構頻繁型樣基底及頻繁型樣池……………………………………28
3.5.2篩選…………..………………………………………………………28
3.5.3探勘指定型樣其出現之頻繁數量及支持度…………………………29
3.5.4產生關聯規則…………………………………………………………29
3.5.5其他輔助支援功能……………………………………………………29
第四章 實驗結果與分析
4.1 第一期 - 探勘一般客戶群……………………………………………31
4.1.1一般客戶群資料的特性………………………………………………31
4.1.2建構頻繁型樣基底……………………………………………………31
4.1.3建構基本頻繁型樣池…………………………………………………37
4.1.4資料項目與出現次數…………………………………………………38
4.1.5依據探勘目標產生頻繁型樣…………………………………………39
4.1.6不同分類給以不同門檻值之限制……………………………………40
4.2 第二期 - 探勘特定客戶群……………………………………………41
4.2.1特定客戶群資料的特性………………………………………………42
4.2.2依據最低支持度門檻值所產生之頻繁型樣………………………….46
4.2.3篩檢 - 指定必須存在分類……………………………………………47
4.2.4篩檢 - 刪除重複型樣及不完整型樣…………………………………48
4.2.5尋找頻繁型樣中之共同分類…………………………………………49
4.2.6頻繁型樣與分類間之關聯規則………………………………………49
4.2.7最低支持度門檻值與關聯規則………………………………………60
4.2.8支持度與信心度………………………………………………………62
4.3 實驗結果討論與分析……………………………………………………65
4.3.1探勘一般客戶群………………………………………………………65
4.3.2探勘特定客戶群………………………………………………………67
第五章 結論及後續研究
5.1 結論與建議………………………………………………………………70
5.2 後續研究與建議…………………………………………………………71
附件 A 最低支持度門檻值與關聯規則之對應關係………………………72
附件 B 支持度與信心度....………………………………………………79
附件 C 資料探勘特定客戶群之關聯法則結果……………………………87
參考文獻

[1] Chang R.Y. & Kelly P.K.(1995). A Practical Guide To Impoving Internal And External Customer Satisfaction, London:Kogan Page.
[2] John E.(2000). Achieving Customer Satisfaction Using Evolutionary Processes, Software Quality Professional,vol.3.
[3] 王伯珣 (2002) ADSL寬頻網路服務消費者行為及市場區隔之研究
[4] 蔡孟恭 (1999) 資料倉儲及資料挖掘應用於電信業之研究
[5] 羅美榮 (2001) 高效率之高頻資料項目型樣挖掘演算法
[6] 邱士軍 (2002) 關聯規則演算法之實作和效能評估.
[7] Jiawei Han, Jian Pei, and Yiwen Yin, “Mining Frequent Patterns without Candidate Generation”. In Proc. 2000 ACM-SIGMOD Int. Conf. Management of Data ( SIG-MOD’00) , Dallas, TX, May 2000, pp. 1-12.
[8] Anurag Srivastava, Eui-Hong Han, Vipin Kumar, Vineet Singh, “Parallel Formulations of Decision-Tree Classification Algorithms”Data Mining and Knowledge Discovery3 (3): 237-261, September 1999
[9] Yi Lin, “Support Vector Machines and the Bayes Rule in Classification”Data Mining and Knowledge Discovery 6 (3): 259-275, July 2002
[10] Johannes Gehrke, Raghu Ramakrishnan, Venkatesh Ganti , “RainForest─A Framework for Fast Decision Tree Construction of Large Datasets”Data Mining and Knowledge Discovery 4 (2-3): 127-162, July 2000
[11] Rajeev Rastogi, Kyuseok Shim, “PUBLIC: A Decision Tree Classifier that Integrates Building and Pruning” Data Mining and Knowledge Discovery 4 (4): 315-344, October 2000
[12] Saharon Rosset, Einat Neumann, Uri Eick, Nurit Vatnik, ”Customer Lifetime Value Models for Decision Support” Data Mining and Knowledge Discovery 7 (3): 321-339, July 2003
[13] S.D. Lee, David W. Cheung, Ben Kao, ”Is Sampling Useful in Data Mining? A Case in the Maintenance of Discovered Association Rules” Data Mining and Knowledge Discovery 2 (3): 233-262, September 1998
[14] Brij Masand, Piew Datta, D.R. Mani, Bin Li, “CHAMP: A Prototype for Automated Cellular Churn Prediction”Data Mining and Knowledge Discovery 3 (2): 219-225, June 1999
[15] J. Sunil Rao, “Bootstrapping to Assess and Improve Atmospheric Prediction Models” Data Mining and Knowledge Discovery 4 (1): 29-41, April 2000
[16] Heikki Mannila, Hannu Toivonen, A. Inkeri Verkamo, “Discovery of Frequent Episodes in Event Sequences” Data Mining and Knowledge Discovery 1 (3): 259-289, 1997
[17] Jef Wijsen, Robert Meersman , “On the Complexity of Mining Quantitative Association Rules” Data Mining and Knowledge Discovery 2 (3): 263-281, September 1998
[18] Craig Silverstein, Sergey Brin, Rajeev Motwani, ”Beyond Market Baskets: Generalizing Association Rules to Dependence Rules” Data Mining and Knowledge Discovery 2 (1): 39-68, January 1998
[19] Brock Barber, Howard J. Hamilton , ”Extracting Share Frequent Itemsets with Infrequent Subsets” Data Mining and Knowledge Discovery 7 (2): 153-185, April 2003
[20] Zhexue Huang , “Extensions to the k-Means Algorithm for Clustering Large Data Sets with Categorical Values” Data Mining and Knowledge Discovery 2 (3): 283-304, September 1998
[21] Johannes Grabmeier, Andreas Rudolph , “Techniques of Cluster Algorithms in Data Mining” Data Mining and Knowledge Discovery 6 (4): 303-360, October 2002
[22] Jeremy Kepner, Rita Kim , “Cluster Detection in Databases: The Adaptive Matched Filter Algorithm and Implementation” Data Mining and Knowledge Discovery 7 (1): 57-79, January 2003
[23] R. Agrawal, R. Srikant, "Mining Sequential Patterns," Proc. of the Int''l Conference on Data Engineering (ICDE), Taipei, Taiwan, March 1995. Expanded version available as IBM Research Report RJ9910, October 1994.