A Review of Studies On Machine Learning Techniques.

International Journal of Computer Science and Security

Pradeep Kumar Bhatia¹  Yogesh Singh¹  Omprakash Sangwan¹ 

[1] $$

关键词: Machine Learning Techniques (MLT);    Neural Networks (NN);    Case Based Reasoning (CBR);    Classification and Regression Trees (CART);    Rule Induction;    Genetic Algorithms and Genetic Programming;   

DOI  :  

来源: Computer Science and Security

【 摘 要 】

This paper provides an extensive review of studies related to expert estimation ofsoftware development using Machine-Learning Techniques (MLT). Machinelearning in this new era, is demonstrating the promise of producing consistentlyaccurate estimates. Machine learning system effectively “learns” how to estimatefrom training set of completed projects. The main goal and contribution of thereview is to support the research on expert estimation, i.e. to ease otherresearchers for relevant expert estimation studies using machine-learningtechniques. This paper presents the most commonly used machine learningtechniques such as neural networks, case based reasoning, classification andregression trees, rule induction, genetic algorithm & genetic programming forexpert estimation in the field of software development. In each of our study wefound that the results of various machine-learning techniques depends onapplication areas on which they are applied. Our review of study not onlysuggests that these techniques are competitive with traditional estimators on onedata set, but also illustrate that these methods are sensitive to the data on whichthey are trained.

【 授权许可】

Unknown