Presented a method for estimating the cost of software using PCA to reduce the size and with the help of data mining

Today s data mining is one of the most important topics. Domain data mining is a combination ofcomputer science and statistics that has been growing over recent years with increasing digital data and thecomputing power of computers is particularly limited. One of the domains of data mining is the category ofsoftware cost estimation. In this paper, the concept of machine learning algorithms and COCOMO model,which is one of the most common models of software cost estimation, are presented, as well as a Principalcomponent analysis approach. In this paper, it is shown that the proposed method is suitable for increasing theaccuracy in estimating the software cost. Subsequently, using this method, the initial dataset was droppeddimension and turned into a new dataset, and the best ones were extracted from the features. Using thismethod, different classification algorithms were evaluated based on different criteria. Ultimately, we haveprovided the truth of our claim, which is based on increasing the accuracy of software cost estimation.

کلیدواژه ها:

Increased accuracy ، Software cost estimation ، Principle components analysis ، Data mining