Volume 4 Issue 5



Abstract : Feature s election is a term standard in data mining to reduce inputs to a manageable size for analysis and processing which also focuses on identifying ir relevant information without affecting the accuracy of the classifier . FS selects a subset of relevant feature s and removes irrelevant and redundant features from the raw data to build a robust learning model . FS is very important,not only because of the curse of dimensionality,but also because of data complexities and the quantities of the data faced by multiple disciplines,such as machine learning,data mining,statistics,pattern recognition and bioinformatics. In recent years,we have seen extensive research in feature selection which has been expanding in depth and in breadth from simple to more advanced techniques,from supervised to unsupervised and semi - supervised feature selection . This paper presents a state - of - art survey of feature selection techniques.

Pages : 1-4

Downloads : 2777

Publication Date :

Modified Date : 2017-05-20

Cite/Export :

KALPANA JAIN , "A SURVEY ON FEATURE SELECTION TECHNIQUES", IJIERT - International Journal of Innovations in Engineering Research and Technology, Volume 4 Issue 5, ISSN : 2394-3696, Page No. 1-4