• C SUDHAKAR REDDY
• DR K VENU GOPAL RAO

In the current digital world huge voluminous data is getting generated and stored by various stake holder of the system. To improve the performance organizations started utilizing this data for analysis for better decision making and improve the performance. Data mining/machine learning techniques are most widely used techniques for analyzing the data. As the real world data is very huge and may attributes are included it has become mandatory to apply some feature selection algorithm for dimensionality reduction in order to identify the features relevant for given type of analysis and use of feature selection algorithm generates a model with better accuracy and reduced learning time and only such features are used for analysis and so that the model will be created with high accuracy and less learning time. In this paper we propose a measure called relative dependency for identifying redundant attributes. Comparative performance of various classification algorithms is illustrated in the paper on various data set collected from UCI repository.

Keywords: Feature Selection, Attribute Dissimilarity

"An Integrated Approach for feature selection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 10, page no.2724-2726, October-2020, Available :http://www.jetir.org/papers/JETIR2010352.pdf

"An Integrated Approach for feature selection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 10, page no. pp2724-2726, October-2020, Available at : http://www.jetir.org/papers/JETIR2010352.pdf