A Threshold fuzzy entropy based feature selection method applied in various benchmark datasets using Ant-miner algorithm

Journal Title: International Journal of Modern Engineering Research (IJMER) - Year 2015, Vol 5, Issue 3

Abstract

Large amount of data have been stored and manipulated using various database technologies. Processing all the attributes for the particular means is the difficult task. To avoid such difficulties, feature selection process is processed.In this paper,we are collect a eight various benchmark datasets from UCI repository.Feature selection process is carried out using fuzzy entropy based relevance measure algorithm and follows three selection strategies like Mean selection strategy,Half selection strategy and Neural network for threshold selection strategy. After the features are selected, they are evaluated using Radial Basis Function (RBF) network,Stacking,Bagging,AdaBoostM1 and Antminer classification methodologies.The test results depicts that Neural network for threshold selection strategy works well in selecting features and Ant-miner methodology works best in bringing out better accuracy with selected feature than processing with original dataset.The obtained result of this experiment shows that clearly the Ant-miner is superiority than other classifiers.Thus, this proposed Antminer algorithm could be a more suitable method for producing good results with fewer features than the original datasets

Authors and Affiliations

P. Devipriya , P. Divya , R. Srivinitha , J. Joshimadevi , S. Saravanan , R. G. Suresh Kumar

Keywords

Related Articles

Some New Operations on Fuzzy Soft Sets

 Fuzzy soft set is one of the recent topics developed for dealing with the uncertainties present in most of our real life situations. The parameterization tool of soft set theory enhances the flexibility of its...

A flywheel energy storage system for an isolated micro-grid

The paper presents an investigation into the effects of integrating a Magnetically Loaded Composite (sMLC) flywheel to an isolated micro-grid. The Fair Isle is a small island located in northern Scotland, and s...

 Suitability of Nigerian Agricultural By-Products as Cement Replacement for Concrete Making

 Abstract: This work investigated the strength characteristics of binary blended cement concrete made with Ordinary Portland Cement (OPC) and each of eight agricultural by-products in South Eastern Nigeria, namely R...

 Implementing Re-Active Power Compensation Technique in Long Transmission System (750 Km) By Using Shunt Facts Control Device with Mat Lab Simlink Tool

 ABSTRACT: This paper attempts to implementation and simulation of the Fuzzy logic control of firing angle for SVC in order to achieve better, smooth and adaptive control of reactive power by using facts technology,...

 Experimental Investigation of Performance and Emission Characteristics of Biodiesel from Sterculia Striata

 With this increase in population, industry, transportation the cost of fossil fuels will grow dramatically. There is a need of new technologies for fuel extraction using feed stocks that do not threaten food securi...

Download PDF file
  • EP ID EP89290
  • DOI -
  • Views 140
  • Downloads 0

How To Cite

P. Devipriya, P. Divya, R. Srivinitha, J. Joshimadevi, S. Saravanan, R. G. Suresh Kumar (2015). A Threshold fuzzy entropy based feature selection method applied in various benchmark datasets using Ant-miner algorithm. International Journal of Modern Engineering Research (IJMER), 5(3), 32-43. https://europub.co.uk/articles/-A-89290