Software Effort Estimation Using Data Mining Techniques
This paper describes an empirical study undertaken to investigate the quantitative aspects of application of data mining techniques to build models for Software effort estimation. The techniques chosen are Multi linear regression, Logistic regression and CART.Empirical evaluation using three fold cross validation procedure has been carried out using three bench marking datasets of software projects, namely, Nasa93, Cocomo81, and Bailey Basili. We observed that: (1) CART technique is suitable for Nasa93 and Nasa93_5. (2). Multiple Linear Regression is suitable for Nasa93_2, Cocomo81s, Cocomo81o and Basili Bailey. (3). Logistic Regression is suitable for Nasa93_1, Cocomo81 and Cocomo81e. It is concluded that data mining techniques tend to help estimating in the best way possible as they are objective and are applicable to unlimited sets of data.
KeywordsSoftware effort estimation CART Logistic regression
Unable to display preview. Download preview PDF.
- 1.Bowie, D.K.: Using multi variate logistic regression analysis to predict black male student persistence at a predominately white institution. An approach investigating the relationship between engagement and persistence. PhD Dissertation. Louisiana State University and Agricultural and Mechanical College, USA (2006)Google Scholar
- 2.Saha, G.: Applying logistic regression model to the examination results data. Journal of Reliability and Statistical Studies 4(2), 1–13 (2011)Google Scholar
- 3.Dejaeger, K., Verbeke, W., Martens, D., Baesens, B.: Data Mining Techniques for Software Effort estimation: a Comparative study. IEEE Transactions on Software Engineering 1(1), 1–25 (2011)Google Scholar
- 4.Mendes, E.: Cost Estimation Techniques for Web Projects. IGI Publishing (2008)Google Scholar
- 6.Brown, S.H.: Multiple linear regression analysis: a matrix approach with matlab. Alabama Journal of Mathematics Spring/Fall, 1–3 (2009)Google Scholar
- 7.Tirimula Rao, B., Sameet, B., Kiran Swathi, G., Vikram Gupta, K., Raviteja, C., Sumana, S.: A Novel Neural Network approach for Software Cost Estimation Using Functional Link Artificial Neural Networks. International Journal of Computer Science and Network Security (IJCSNS) 9(6), 126–131 (2009)Google Scholar