臺灣博碩士論文加值系統

在醫院的醫療資訊環境中不斷有各類型軟體專案的需求。能有效的預估軟體開發時程，對醫院的資訊單位在執行成效上有極大的影響，目前最大的問題在於評估軟體專案時，沒有相對應的一套評估流程，以適切的評估軟體的工時，以及人員的時程安排。因此找出一個有效的軟體時程預估之流程與作法，對醫院的資訊單位而言有其必要性。
本研究以南部某醫學中心資訊室為例，提出經驗式的評估工時的方法。首先收集以往的25筆專案資料，以此資料集為基礎，藉由功能點分析，以迴歸分析預測每個功能點數將需要撰寫的程式碼數量，作為COCOMO II的輸入參數。再以COCOMO II的估算工作量公式透過對數變換，取得案例單位的經驗參數，再以COCOMO II非線性迴歸模型配合專案的權重與成本因子，作專案時程的預估，在實作過程，我們發現若直接以簡單迴歸模式來預估，在檢驗迴歸模型時，會有解釋能力不足的問題，其原因為開發架構與人員的不同，造成使用功能點分析評估軟體複雜度時的差異。因此再以多元迴歸分析來發掘關鍵變數對於迴歸模型的預測影響，以改進預測解釋能力的問題。其次我們請4位資深同仁，就這些專案進行風險評估，我們將風險分為5個層級，再以資料探勘、類神經網路、決策樹等方式，根據專案風險與預估誤差之間的關聯性，修正相對應的誤差，讓預估更加準確，最後再加入11筆最新的專案資料作為驗證。目前相關研究只使用類神經網路及決策樹的方式，評估結果顯示，對於案例單位的專案資料集來說，資料探勘的修正方法，平均預估穩定度較類神經網路為高，未來將以此模型作為案例單位新專案開發之參考。

In this research, we propose a refined software cost prediction model based on the experiences of the department of medical informatics of a medical center in southern Taiwan. The model is an extension of COCOMO II. Firstly, we analyze 25 completed projects. We find that the single regression model is not good enough to interpret the measured data. We then introduce the multi-variable regression model. After deriving the cost model, we introduce the risk factor by the help of four senior staffs to define the risk levels of the projects. We divide the risk factor into 5 levels. Currently, related researches fine tune the results using decision tree and neural network approaches. We propose to use association rules (of data mining).
We use 11 newly developed projects to evaluate the proposed approach.The results show that association rules and neural network approaches would improve the accuracy of estimation model and they are better then the decision tree approach. The association rules approach shows 13% improvement of MMRE(Mean Magnitude of Relative Error), 35% improvement of PRED(Prediction level)(25%), while the neural network approach shows 14% improvement of MMRE, 31% improvement of PRED(25%). In comparison, association rules approach is better than neural network. Hence, we suggest to use the association rules approach in the future.

第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 2
1.3 論文架構 3
第二章 文獻探討 5
2.1 工時預估模型相關研究 5
2.2 軟體工時估算方法 7
2.3 軟體規模評估方法 9
2.4 成本預估模型 15
2.5 工時預估模型的改善方法 21
2.6 工時預估模型的量化評估 23
第三章 工時預估模型設計 25
3.1 工時預估流程 25
3.2 風險評估及誤差等級 27
3.3 案例公司概述 28
3.4 工時預估執行過程 30
第四章 專案資料驗證結果與分析 52
4.1 資料來源 52
4.2 驗證方法 52
4.3 原始功能點構面與修改構面驗證 53
4.4 一般化與專案分類模型驗證 54
4.5 資料探勘-關聯法則規則庫 55
4.6 關聯法則信心度驗證 58
4.7 相關誤差修正方法驗證 59
4.8 選擇預估修正方法 63
第五章 結論與未來研究方向 67
5.1 結論 67
5.2 未來研究方向 68
參考文獻 70

[1]A. Abran and P.N. Robillard, Function analysis: an emprirical study of its measurement process, IEEE Transactions on Software Engineering, vol. 22, no. 12, p895-910, December 1996.
[2]Norman E. Fenton, “Software metrics: theory, tools and validation, Software Engineering Journal, vol. 2, no. 1, p65-84, January 1990.
[3]M. H. Halstead,Elements of software science, Elsevier North-Holland, New York, 1977.
[4]Edilson J.D. Candido and R. Sanches, Estimating the size of web applications by using a simplified function point method, Proc. Web Media and Latin American Web Congress, p98-105, 2004.
[5]A.J. Albrecht and J. Gaffney, “Software function, source lines of code, and development effort prediction: a software science validation, IEEE Transactions on Software Engineering, vol.9, no.6, p639-648, 1983.
[6]DoD(USA), “Department of Defense handbook work breakdown structure, chapter 1, p4, 1998.
[7]B. Boehm, Software engineering economics, Englewood cliff, NJ: Prentice-Hall, 1981.
[8]International Function Point Users Group, Function point counting practices manual release 4.3.1, 2010.
[9]Ray Boehm, “Frequently asked questions, International Function Point Users Group, 2002.
[10]L. Wallace, M. Keil, A. Rai, “Understanding software project risk：a cluster analysis, Information ＆ Management, vol. 42, issue 1, p115-125, 2004.
[11]B. Boehm, Software engineering economics, Prentice Hall PTR; 1st edition, 1981.
[12]J.E. Matson, B.E. Barrett, J.M. Mellichamp, “Software development cost estimation using function points, IEEE Transactions on Software Engineering, vol. 20, no. 4, p275–287, 1994.
[13]S.-J. Huang, C.-Y. Lin and, N.-H. Chiu, “Fuzzy decision tree apporoach for embedding risk assessment information into software cost estimation model, Software Engineering and Software, vol. 22, no. 2, p297-313, 2006.
[14]S. Weisberg, Applied linear regression, Wiley, New York, 1985.
[15]R. Agrawal,T. Imielinski and A. Swami , “Mining association rules between sets of items in large databases, Proc. the 1993 ACM SIGMOD international conference on Management of data, Washington, D.C., United States, p207-216, 1993.
[16]R. Agrawal and R. Srikant, “Fast algorithms for mining association rules, Proc. the 20th International Conference on Very Large Data Bases(VLDB), Santiago, p487-499, 1994
[17]B. Boehm, E. Horowitz, R. Madachy, D. Reifer, B. Clark, B. Steece, W. Brown,S. Chulani, and C. Abts, Software cost estimation with COCOMO II, Englewood cliff, NJ: Prentice-Hall, 2000.
[18]C. E. Walston and C. P. Felix, “A method of programming measurement and estimation, IBM Systems Journal, vol. 16, no. 1, p54–73, 1977.
[19]John W. Bailey and Victor R. Basili, “A meta-model for software development resource expenditures, Proceedings of the 5th international conference on Software engineering, USA, p107–116, 1981.
[20]A.J. Albrecht and J.E. Gaffney, “Software function,source lines of code, and development effort prediction, IEEE Transactions on Software Engineering, vol. 9, no. 6, p639–648, 1983.
[21]C. Jones, “Software cost estimation in 2002, The Journal of Defense Software Engineering, vol. 15, no. 6, p4-8, 2002.
[22]P. Jodpimai, P. Sophatsathit, and C. Lursinsap, “Analysis of effort estimation based on software project models, Proc. Communication and Information Technology, IEEE Press, p715-720, 2009.
[23]D. Port and M. Korte, “Comparative studies of the model evaluation criterions MMRE and PRED in software cost estimation research, Proc. ACM-IEEE international symposium on Empirical software engineering and measurement, ACM Press, New York, USA, p51-60, 2008.
[24]J. J. Cuadrado-Gallego, M.-A. Sicilia, M. Garre, and D. Rodriguez, “An empirical study of process-related attributes in segmented software cost-estimation relationships, The Journal of Systems and Software, vol. 79, p353–361, 2006.
[25]Attarzadeh I. and Siew Hock Ow, “A novel soft computing model to increase the accuracy of software development cost estimation, Proc. The 2nd International Conference on Computer and Automation Engineering, Singapore, p603-607, 2010.
[26]SPSS 17,http://www.spss.com/ (last visited:Dec 15th,2010)
[27]Weka 3, http://www.cs.waikato.ac.nz/ml/weka/ (last visited:Jan 16th,2011)
[28]Super PCNeuron 5.0, http://people.chu.edu.tw/~icyeh/spcn/ (last visited:Jan 20th,2011)
[29]榮思哲，以WBS為基礎之人力估算模型與功能點分析服務，國立中央大學資訊工程學系碩士論文，2007。
[30]陳昱祥，基於COCOMO II之元件化軟體開發成本預估，國立成功大學資訊工程學系碩士論文，2009。
[31]林煒佑，使用類推及迴歸技術之同質性軟體維護工時評估系統，國立臺灣海洋大學資訊工程學系碩士論文，2007。
[32]林榮泰，應用模糊決策樹分析於研發型專案風險之評估，國立成功大學製造工程研究所碩士論文，2007。
[33]湯佳天，問題單處理與專案派工系統之整合與實作，中原大學資訊工程研究所，2009。