A literature review of feature selection techniques and applications: Review of feature selection in data mining

Water is the elixir of life. It is a vital component of human survival. Water should be purified for better and healthy style life of all living and non-living things. The quality of water plays an important role for all living beings. Water used for drinking purpose should be colourless, odourless and free from excess salts. Detecting such a variety of contamination from the drinking water becomes a challenging task. Feature selection acts as a significant role in identifying irrelevant features and redundant features from large dataset. Feature selection is a preprocessing course of action universally used for large amount of data. Feature selection concepts instruct us, to pick a subset of features or catalog of attribute or variables which helps to build an efficient model for describing the selected subset. Other than selecting the subset, it also congregate some other purposes, such as dimensionality reduction, compact the amount of data which are required for learning process, progress in predictive accuracy and increasing the constructed models. The main aim of this work is to investigate about the concept of feature selection, various criterions of feature selection methods and some existing methods are discussed from 1997 till 2014 and address the issues and challenges of feature selection.