High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

A Neuro-Genetic Approach for Detecting Suspicious Criminal Activities on Emails

Author(s):

Hitesh Mahanand , Mats school of Engineering ⁢ Deepak Kumar Xaxa, Mats school of Engineering &IT

Keywords:

Classification Techniques, Criminal Activities, Feature Selection, Genetic, Hybrid, Neuro, Suspicious

Abstract

The paper presents a hybrid data mining classification technique for detection of suspicious criminal activities on e-mails. In the paper we proposed a Neuro-Genetic approach in which, a Genetic Algorithm is used for feature reduction & selection and Neural Network with back-propagation is used for classification. The presented hybrid model focuses on the evaluation of knowledge based learning’s to detecting the suspicious criminal activities on e-mails. In the literature, various machine learning algorithms achieved good accuracy for the related task. We have identified the combination of those algorithms results that improve the performance of the existing algorithms.

Other Details

Paper ID: IJSRDV3I50503
Published in: Volume : 3, Issue : 5
Publication Date: 01/08/2015
Page(s): 669-672

Article Preview

Download Article