The data dimensionality reduction and features weighting in the classification process using forest optimization algorithm

The paper presents the data dimensionality reduction in the classification process, with a special presentation of using the ability of features weighting by determining the level of importance of a given attribute in the data vector. This reduction was implemented using the Forest Optimization Algorithm (FOA) and the use of a classifier allowing to enter the importance of each attribute for a data vector. The paper presents both, a description of the capability of using the FOA algorithm as well as the possibility of introducing modifications which allows to regulate the objective function between the obtained classification result and the number of reduced features. The conducted tests and obtained results were also presented. At the end of paper, a summary and the final conclusions are provided.