如何更精確的預估開發所需的軟體工作量對軟體專案的規劃、監控與成功與否，是一個非常重要的關鍵。因為如果低估軟體專案開發的工作量，將會因為資源分配不足，而造成軟體品質不佳或導致專案失敗；反之若高估軟體專案開發的工作量，則會因投入過多的資源而造成資源的浪費。以往文獻中主要探討單一因子資料分群對所建構之軟體工作量預估模式精確性的影響，而未探討以多個因子作資料分群所建構的軟體工作量預估模式精確性的影響，因此本研究的主要目的是探討資料同質性越高，所建構出軟體工作量預估模式的精確性是否有改善的問題。我們以資料探勘的技術來對軟體專案歷史資料以多個因子與單一因子來分群，再以類神經網路與統計迴歸模式來建構個別的預估模式，進而比較多個因子、單一因子與未分群之資料分群所建構出軟體工作量預估模式之精確性。研究結果主要發現為：(1)在非監督式分群情況下，使用單因子和多因子分群所建構之軟體工作量預估模式的精確性都比資料未分群所建構之模式較佳，而監督式分群精確度並無顯著提昇，因此我們認為非監督式分群可以使得資料集比監督式分群具有較高同質性，使得非監督式分群工作量預估精確度較為準確；(2)不論非監督式分群或監督式分群，以類神經網路所建構的軟體工作量預估模式之精確性都比迴歸分析預估模式較佳。

How to estimate software effort more accurately is always a key to the success of a software project. Under-estimating the effort needed for software development may cause the sacrifice of the software quality and seriously lead to the failure of the software development project because of insufficient distribution of the allocated resources. However, over-estimating the software development effort may also cause the problem of the inefficient usage of allocated resources and further lose the chance of gaining the software project in the price bidding because of allocating too many resources. To date, the researches on the effect of the accuracy of software effort estimates by clustering software project data have only concentrated on a single software effort driver. This paper aims at investigating the effect of accuracy of software effort estimation model that is built by using the homogenous software project data. By clustering the project data with single and multiple software effort drives, we use Neural Network and Regression Analysis methods to construct the individual software effort estimation models, and then compare their accuracies with supervised clustering and unsupervised clustering methods of data mining. The results show that: (1) the accuracies of both single and multi-driver software effort estimation models are better than the unclustered model with unsupervised clustering data, but supervised clustering data does not. Therefore, we consider that the unsupervised clustering method has more homogenous than supervised clustering method. (2) No matter if the software project data are clustered with supervised learning or unsupervised learning, the accuracy of Neural Network software effort estimation model is better than the Regression Analysis model.

[1] 王進德，蕭大全，「類神經網路與模糊控制理論入門」，全華科技圖書股份有限公司印行，2000。
[2] 林傑毅，「整合風險評估之軟體專案預估模式」，碩士論文，國立臺灣科技大學資訊管理研究所，民國91年
[3] 吳典璋，「以類神經網路預估軟體開發成本之研究」，碩士論文，國立中央大學資訊工程研究所，民國87年。
[4] 孫文祥，「談軟體專案失敗的六大關鍵」，民國89年。
[5] 葉怡成，「類神經網路模式應用與實作」，儒林圖書有限公司，民國92年。
[6] 張紹勳、張紹評、林秀娟，「SPSS For Windows統計分析─初等統計與高等統計(下冊)」，松崗電腦圖書公司，民國91年。
[7] 楊維寧，「統計學」，新陸出版，初版，民國92年。
[8] 劉亦真，「軟體專案資料分類對工作量預估值精確性之影響」，碩士論文，國立台灣科技大學資訊管理研究所，民國92年。
[9] 鄭宇修，「模糊類神經網路軟體工作量預估模式」，碩士論文，國立台灣科技大學資訊管理研究所，民國91年。
[10] Albrecht, A.J. and Gaffney, J.E., “Software Function, Source Lines of Code, and Development Effort Prediction: A Software Science Validation,” IEEE Transactions on Software Engineering, Vol. 9, No. 6, pp. 639-648, 1983.
[11] Angelis, L., Stamelos, I. and Morisio, M., “Building a Software Cost Estimation Model Based on Categorical Data,” Pro. of the Seventh International Software Metrics Symposium METRICS 2001, London, pp.4-15, April 2001.
[12] Bode, J., “Decision support with neural networks in the management of research and development: concepts and application to cost estimation,” Information and Management, Vol.34, pp.33-40, March 1998.
[13] Boehm, B.W., “Software Cost Estimation with COCOMOⅡ,” Prentice Hall PTR,2000.
[14] Boehm, B.W., “Software Engineering Economics,” Englewood cliff, NJ:Prentice-Hall, 1981.
[15] Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J., “Classification and Regression Tree,” Wadsworth, 1984.
[16] CMMI Product Team, “Capability Maturity Model Integration v1.1,” CMMI-SW/SE/IPPD/SS, Staged representation CMU/SEI-2002-TR-011.March 2002.
[17] Conte, S.D., Dunsmore, H.E., Shen, V. Y., “Software Engineering Metrics and Models,” MenloPark, CA: Benjamin/Cummings, 1986.
[18] Chris Lokan, Terry Wright, Peter R.Hill and Michael Stringer, “Organizational Benchmarking Using the ISBSG Data Repository,” IEEE Software, Sep/Oct 2001.
[19] Dolado, J.J., “A Study of the Relationships among Albrecht and Mark II Function Points, Lines of Code 4GL and Effort.” The Journal of Systems and Software, Vol.37, pp.161-173, 1997.
[20] Ferens, Daniel V. and Gurner, Robert B., “An evaluation of three function point models for estimation of software effort,” Institute of Electrical and Electronics Engineers, pp.635-642, 1992.
[21] Finnie, G.R., Wittig G.E. and Desharnais J-M., “A Comparison of software effort estimation techniques: using function points with neural networks, case-based reasoning and regression model,” Systems software, Vol.39, pp.281-289,1997.
[22] Han, J. and Kamber, M., “Data Mining: Concepts and Techniques,” Elsevier Pte Ltd., 2000.
[23] Heiat, A., “Comparison of artificial neural network and regression models for estimating software development effort,” Information and Software Technology, Vol.44, pp.911-922, 2002.
[24] Hughes, R.T., “Expert Judgement as an Estimating Method,” Information and Software Technology, Vol. 38, pp.67-75, 1996.
[25] International Software Benchmarking Standards Group, ISBSG, http://www.isbsg.org
[26] ISBSG, The Benchmark-Release8, International Software Benchmarking Standards Group, Warrandyte, Australia, 2004.
[27] Jeffery, R.; Ruhe, M. and Wieczorek, I., “Using Public Domain Metrics to Estimate Software Development Effort,” Pro. of the Seventh International Software Metrics Symposium METRICS 2001, London, pp.16-27, April 2001.
[28] Jeffery, R., Ruhe, M. and Wieczorek, I., “A Comparative Study of Two Software Development Cost Modeling Techniques using Multi-organizational and Company-specific Data,” Proceedings of the ESCOM 2000, pp.239-247, March 2000.
[29] Jørgensen, M., “A review of studies on expert estimation of software development effort,” Journal of Systems and Software, Vol.70, pp.37-60, 2004.
[30] Kemerer, C.F., “An empirical validation of software cost estimation models,” Commun. ACM, Vol.30, no.5, pp.416-429, 1987.
[31] Kitchenham, B.A., Pfleeger, S.L., McColl, B. and Eagan, S., “An empirical study of maintenance and development estimation accuracy,” Journal of Systems and Software, Vol.64, pp.57-77, 2002.
[32] Lokan, C.; Wright, T.; Hill, P.R. and Stringer, M., “Organizational Benchmarking Using the ISBSG Data Repository,” IEEE Software, pp. 26-32, September/October 2001.
[33] Low, G.C. and Jeffery, D.R., “Function points in the estimation and evaluation of the software process,” IEEE Transactions on Software Engineering, Vol.16, pp.64-71, 1990.
[34] Mair, C., Kadoda, G.., Lefley, M., Phalp, K., Schofield, C., Shepperd, M. and Webster, S., “An investigation of machine learning based prediction systems,” Journal of Systems and Software, Vol. 53, pp. 23-29, 2000.
[35] Maxwell, K., Van Wassenhove, L. and Dutta, S., “Performance Evaluation of General and Company Specific Models in Software Development Effort Estimation,” Management Science, Vol.45,No.6, pp.787-803, June 1999.
[36] Matson, J.E.; Barrett, B.E.; Mellichamp, J.M., “Software development cost estimation using function points,” IEEE Transactions on Software Engineering, Vol.20, pp.275-287, April 1994.
[37] Pengelly, A., “Performance of effort estimating techniques in current development environments,” IEEE Transactions on Software Engineering, Vol.10, pp.162-170, Sept 1995.
[38] Rask, R., Laamanen, P. and Lyytinen, K., “Simulation and Comparison of Albrecht's Function Point and Demarco's Function Bang Metrics in a CASE Environment,” IEEE Transactions on Software Engineering, Vol.19, pp.661-671, 1993.
[39] Ryder, J., “Fuzzy Modeling of Software Effort Prediction,” IEEE Information Technology Conference, 1998.
[40] Shepperd, M. J. and Schofield, C., “Estimating Software Project Effort Using Analogies,” IEEE Transactions on Software Engineering, Vol.23, pp.736-743, 1997.
[41] Shepperd, M.J. and Kadoda, G.F., “Comparing Software Prediction Techniques Using Simulation,” IEEE Transactions on Software Engineering, Vol.27, pp.1014-1022, 2001.
[42] Sommerville, L., “Software Engineering,” Forth Edition, Addision Wesley, 1995.
[43] Srinivasan, K. and Fisher, D., “Machine Learning Approaches to Estimation Software Development Effort,” IEEE Transactions on Software Engineering, Vol.21, pp.126-137, February 1995.
[44] Symons, C.R., “Function point analysis: Difficulties and improvements,” IEEE Transactions on Software Engineering, Vol.14, pp.2-11, Jan 1988.
[45] The Standish Group, “The CHAOS Report (1994)”,
http://www.standishgroup.com/sample_research/index.php