臺灣博碩士論文加值系統

本研究的目的在於對群聚演算法的研究與探討，並提出群聚特性評估與群聚維度選擇的方法，協助使用者了解群聚參數與資料維度對於群聚的影響與意義，藉此強化群聚演算法的使用。在以往的研究中，已經有釵h學者提出各種類型的群聚演算法，這些演算法大多都需要設定群聚參數，且參數的選擇會影響群聚後的結果，因此，使用者必須要充分了解參數對於群聚的意義並選擇適當的群數參數，才能有效的利用群聚演算法來幫助解決決策上的問題，基於以上因素，本研究將對於群聚的資料分佈意義以及參數對群聚的意義與各群聚間的關係做更進一步的分析與描述，並提出一個新的群聚評估式，以期協助決策者有效地找出適當的群聚參數。另一方面，我們由部分維度群聚演算法的文獻探討中，知道若使用全部維度進行群聚，不僅增加計算上的複雜程度，而使用者也很難理解所得到的群聚所代表的意義，因此，我們將提出一個重要維度選擇的方法，以較少的資料維度來取代全部的資料維度進行群聚分析，並將分析哪些維度屬性對於資料有較強的解釋能力。
最後，我們發現若將全維度及部分維度群聚演算法配合使用，可以得到更多的資訊。我們特別舉目標行銷的應用為例，提出一個維度比較的演算法來幫助決策者從客戶交易的資料中找出潛在的行銷對象。相信加入部分維度的群聚資訊，群聚演算法可以有更廣泛的應用。

The goal of our research is to study the clustering algorithms and to analyze the characteristics of clusters for the decision of the parameters of the clustering algorithms. In the past researches, all kinds of the clustering algorithms are proposed for dealing with high dimensional data in large data sets. Nevertheless, almost all of clustering algorithms require input parameters while the users do not have enough domain knowledge to determine the input parameters. Thus, our researches will analyze and discuss the meaning of data distributions in the cluster and the meaning about cluster parameters. We propose a formula for evaluating the important factors in a cluster to determine whether the cluster is well clustered. Moreover, a new method to find the important dimensions(fields) in database is proposed. It helps the users to mine the information in partial dimensions of the data rather than the full dimensions.
Finally, we find that the clustering algorithms will work well when combine with both the results at full dimensions and partial dimensions. We take as example by applying a new algorithm to the application of target market. The algorithms analyze the consumer's consumption patterns for the decision makers to find the potential consumers and to select the better marketing strategies in order to succeed and survive in the competitive business environment.

中文摘要 I
Abstract III
致謝 IV
目錄 V
圖目錄 VII
表目錄 X
壹、 緒論 1
1-1研究動機 1
1-2 研究目的 2
1-3 研究進行步驟 3
1-4 論文架構 3
貳、文獻回顧 5
2-1 切割式群聚演算法 6
2-2 階層式群聚演算法 7
2-3 密度基礎群聚演算法 12
2-4 格子基礎群聚演算法 15
2-5 群聚演算法比較與參數分析 16
參、群聚特性之研究 24
3-1 相關知識 25
3-2 群聚評估 29
3-2-1 鑑別率（Discrimination rate） 30
3-2-2 凝聚率（Agglomerate rate） 34
3-2-3 密度率（Density rate） 38
3-3 實驗與結果 40
肆、維度選擇 49
4-1 部分維度聚群聚演算法 49
4-2 資料維度之選擇 53
4-3 實驗與結果 60
伍、部份維度演算法與目標行銷 65
5-1 目標行銷 66
5-2 新維度比較群聚演算法 68
5-3 應用於目標行銷之實驗範例 71
陸、 結論與未來研究方向 76
6-1 結論.... 76
6-2 研究成果與貢獻 77
6-3 未來研究方向 78
柒、參考文獻 81

[1]Ankerst M., Breunig M., Kriegel H.P. and Sander J., “OPTICS: Ordering points to identify the clustering structure,” In Proc. 1999 ACM-SIGMOD Int. Conf. Management of Data (SIGMOD’99), pp. 49-60, Philadelphia, PA, June 1999.
[2]Agrawal R., Gehrke J., Gunopulos D. and Raghavan P., “Automatic Subspace Clustering of High Dimensional Data for Data Mining Applications,” Int. Conf. Management of Data, pages94-105, Seattle, Washington, 1998.
[3]Agrawal R., Mannila H., Srikant R., Toivonen H. and Verkamo A. I., “Fast discovery of association rules,” Advances in Knowledge Discovery and Data Mining, chapter12, pp. 307-328,AAAI/MIT Press, 1995.
[4]Alfred, S.B., “Market segmentation by personal values and salient product attributes,” Journal of Advertising Research, Vol. 21, No. 1, pp. 29-35, 1982.
[5]Cheng C. H., Fu A.W. and Zhang Y., ”Entropy-based subspace clustering for mining numerical data,” Int. Conf. Knowledge Discovery and Data Mining (KDD’99), pp. 84-93, San Diego, CA, USA, August 1999.
[6]Dash M., Liu M. and Xu X., “`1+1>2': Merging Distance and Density Based Clustering,” In Proc. 7th Int. Conf. Database Systems for Advanced Applications (DASFAA'01), pp. 18-20, Hong Kong, April 2001.
[7]Ester M., Kriegel H.P., Sander J., Wimmer M. and Xu X., “Incremental Clustering for Mining in a Data Warehousing Environment,” In Proc. 24th Int. Conf. Very Large Databases (VLDB'98), pp. 24 - 27, New York City, NY, USA, Aug. 1998.
[8]Ester M., Kriegel H.P., Sander J. and Xu X., “Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise,” In Proc. 1996 Int. Conf. Knowledge Discovery and Data Mining (KDD’96), pp. 226-231, Portland, OR, Aug. 1996.
[9]Estivill-Castro V. and Lee I., “AMOEBA: Hierarchical Clustering Based on Spatial Proximity Using Delaunay Diagram,” In Proc. 9th Int. Spatial Data Handling (SDH2000), pp. 10-12, Beijing, China, Aug. 2000.
[10]Estivill-Castro V. and Lee I., “AUTOCLUST: Automatic Clustering via Boundary Extraction for Massive Point Data Sets,” In Proc. 5th Int. Conf. Geo-Computation, pp. 23-25, University of Greenwich, Kent, UK. Aug. 2000.
[11]Guha S., Rastogi R. and Shim K., “CURE: An efficient clustering algorithm for large databases,” In Proc. 1998 ACM-SIGMOD Int. Conf. Management of Data (SIGMOD’98), pp. 73-84, Seattle, WA, June 1998.
[12]Guha S., Rastogi R. and Shim K., “ROCK: A Robust Clustering Algorithm For Categorical Attribute,” In Proc. 1999 Int. Conf. Data Engineering (ICDE’99), pp. 512-521, Sydney, Australia, Mar. 1999.
[13]Han J. and Kamber M., “Data Mining: Concepts and Techniques,” Morgan Kaufmann, 2000.
[14]Huang Z., "Extensions to the k-Means Algorithm for Clustering Large Data Sets with Categorical Values," Data Mining and Knowledge Discovery, Vol. 2, pp. 283-304, (1998).
[15]Karypis G., Han E.H. and Kumar V., “CHAMELEON: Hierarchical Clustering Using Dynamic Modeling,” IEEE Computer, Vol. 32, No. 8, pp. 68-75, (1999).
[16]Kaufman L. and Rousseeuw PJ, “Finding Groups in Data: an Introduction to Cluster Analysis,” John Wiley & Sons, 1990.
[17]MacQueen J., “Some Methods for Classification and Analysis of Multivariate Observations,” In Proc. 5th Berkeley Symp. Math. Stat. and Prob., Vol. 1, pp. 281-297, 1967.
[18]Ng R. and Han J., “ Efficient and Effective Clustering Method for Spatial Data Mining,” In Proc. 1994 Int. Conf. Very Large Databases (VLDB’94), pp. 144-155, Santiago, Chile, Sept. 1994.
[19]Pride W.H. and Ferrell O.C., “Marketing,” Ninth Edition, Houghton Mifflin, 1995.
[20]Sheikholeslami G., Chatterjee S., and Zhang A., “WaveCluster: A multi-resolution clustering approach for very large spatial databases,” In Proc. 1998 Int. Conf. Very Large Databases (VLDB’98), pp. 428-439, New York, Aug. 1998.
[21]Sander J., Ester M., Kriegel H.P. and Xu X., “Density-Based Clustering in Spatial Databases: The Algorithm GDBSCAN and its Applications,” In Proc. Int. Conf. Data Mining and Knowledge Discovery, Kluwer Academic Publishers, Vol. 2, No. 2, 1998.
[22]Wang W., Yang and Muntz R., “STING: A Statistical Information grid Approach to Spatial Data Mining,” In Proc. 1997 Int. Conf. Very Large Data Bases(VLDB’97), pp. 186-195, Athens, Greece, Aug. 1997.
[23]Wendell, R.S., “Product differentiation and market segmentation as alternative marketing strategies,” Journal of Marketing, Vol. 21, pp. 3-8, 1956.
[24]Xu X., Ester M., Kriegel H.P. and Sander J., “A distribution– based Clustering Algorithm for Mining in Large Spatial Databases," In Proc. 14th Int. Conf. Data Engineering (ICDE'98), Orlando, Florida, USA, 1998.
[25]Zhang T., Ramakrishnan R. and Livny M., “BIRCH: An Efficient Data Clustering Method for Very Large Databases,” In Proc. 1996 ACM-SIGMOD Int. Conf. Management of Data (SIGMOD’96), pp. 103-114, 1996.
[26]魏啟林 編譯、Philip Kotler著，「行銷學精論」，華泰書局、臺北(1988)。
[27]黃志文 著，「行銷管理」，華泰書局、臺北(1995)。