Hybrid Clustering Algorithms for Crime Pattern Analysis
Xavier Alphonse Inbaraj, A. Seshagiri Rao · 2018 International Conference on Current Trends towards Converging Technologies (ICCTCT) · 2018
Data mining can be used to model crime detection problems. Crimes are a social nuisance and cost our society dearly in several ways. Any research that can help in solving crimes faster will pay for itself. About 10% of the criminals commit about 50% of the crimes. Here we look at use of clustering algorithm for a data mining approach to help detect the crimes patterns and speed up the process of solving crime. We will look at Clustering techniques with some enhancements to aid in the process of identification of crime patterns. Without priori knowledge, determine the favorable condition of cluster algorithm imply problems. In this paper we propose a new two-level clustering algorithms and describing evaluation of the numbers of clusters which is involved to the validity of cluster and its stability by the way of learning. Before apply new two level algorithm, preprocessing cluster documentation has done. The main advantage of our proposed algorithm which is not limited and restricted to convex cluster. While compared to the conventional clustering methods, it can recognize illogically shaped clusters. The cost of this algorithm is better- quality to standard two-level clustering methods such as Affinity Propagation(AP) and RBF network and Affinity Propagation (AP) using costed approach and RBF network. The costed approach proves better than traditional Affinity Propagation.