智慧型演算法於評估軟體可靠度成長模型參數之研究

A Study on Evolutionary Computation for Parameter Estimation of Software Reliability Growth Model

張筑婷(Chu-Ting Chang)

指導教授 ： 林金城

大同大學/電機資訊學院/資訊工程學系所/碩士(2010年)

摘要

在軟體工程的領域中，準確地預測軟體可靠度的成長情形非常重要。在過去二十年裡，軟體可靠度成長模型(Software Reliability Growth Model, SRGM)以非齊次蒲瓦松程序(Non-homogeneous Poisson Process ,NHPP)為基礎的模型最為受到重視。近幾年的軟體可靠度預估漸漸導入機器學習的演算法，如決策樹(Decision Tree)、最近鄰居法(K-nearest Neighbor Algorithm)，近期則有類神經網路大量的應用於研究。
本研究預期將針對目前較容易實作的兩種演算法：基因演算法(Genetic Algorithm)、差分演算演化法(Differential Evolution Algorithm)發展一套可以預估軟體可靠度的系統，期望藉由系統可以使得軟體專案管理者對軟體發展時的規劃作更好的評估。

關鍵字

基因演算法 ； 差分演化演算法 ； 軟體可靠度

並列摘要

In the field of software engineering, to predict the software reliablity is very important. In the recent two decades, Software Reliability Growth Model (SRGM) is focus on Non-homogeneous Poisson Process (NHPP) model. Recent years, we use Machine Learning algorithm, like Decision Tree, K-nearest Neighbor Algorithm ,to predict the reliability of software, and recent period, Artificial Neural Network (ANN) is often used.
This research will use Genetic Algorithm (GA) and Differential Evolution Algorithm (DE) to predict software reliability. Through the system, we hope we can get more accretive result of software reliability to help the software project manager.

並列關鍵字

software reliability ； differential evolution algorithm ； genetic algorithm

參考文獻

[Chi09] Kuei-Chen Chiu, "A Study of Software Reliability Growth from the Perspective of Learning Effects", Unpublished doctoral dissertation, National Cheng Kung University

[Geo79] A.L. Goel and K. Okumoto, “A Time Dependent Error Detection Model for Software Keliability and Other Performance Measures,” IEEE Trans. Reliability, Aug. 1979, pp. 206-211.

[Yam83] S. Yamada, M. Ohba, and S. Osaki, “S-Shaped Reliability Growth Modeling for Software Error Detection,” IEEE Trans. Reliability, Dec. 1983, pp. 475-484.

[Gol89] Goldberg, D.E. “Genetic Algorithm in Search.” Optimization and Machine Learning, Addision-Wesley, New York, 1989.

[Sto97] Storn R. and Price, K. “Differential evolution — A simple and efficient heuristic for global optimization over continuous spaces,” Journal of Global Optimization, Vol. 11, No. 4, pp. 341–359, 1997.