Hybrid Technique Based on Clustering for Crime Detection in Data Mining
Keywords:
Crime detection, clustering, classification, Expectation Maximization.Abstract
Crime Detection mainly performed in the Data Mining (DM) to detect the crime efficiently. Crimes are a
social annoyance and charge our society extremely in numerous behaviors. Any research that can facilitate in explaining
crimes quicker will pay for itself and regarding of this, there are many criminals commit. In the present paper, they
implemented a procedure for the design and implementation of crime detection and criminal identification for Indian
cities using DM techniques. Clustering is a most vital field of data analysis and data mining application. It is a set of
methodologies for creating high superiority clusters and high intra-cluster similarity and low inter-class similarity. In
clustering, there is mixture of algorithms to break up the data into groups. K-means is the easiest and most frequently
used algorithm for partitioning the data among the clustering algorithms in the field of scientific and industrial software.
Fuzzy C-means clustering is utilized to amass the data into groups by characterizing certain degree. We used Fuzzy Cmeans and ACO in our proposed work to improve the crime detection rate of different cities.
