Comparative Study of Clustering Algorithms Used in CounterTerrorism
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2014, Vol 16, Issue 6
Abstract
Abstract: Data mining can be used to model crime detection problems, detect unusual patterns, terroristactivities and fraudulent behaviour. We will look at k-means clustering with some enhancements to aid in theprocess of identification of crime patterns. The k-means algorithm is one of the frequently used clusteringmethod in data mining, due to its performance in clustering massive data sets. The final clustering result of thek-means clustering algorithm greatly depends upon the correctness of the initial centroids, which are selectedrandomly. The original k-means algorithm converges to local minimum, not the global optimum. Manyimprovements were already proposed to improve the performance of the k-means, but most of these requireadditional inputs like threshold values for the number of data points in a set. In this paper a new method isproposed for finding the better initial centroids and to provide an efficient way of assigning the data points tosuitable clusters with reduced time complexity
Authors and Affiliations
Sanjay Dwivedi, MCA, MPhil, , Prabhat Pandey, PhD, DSc. OSD, , Manmohan Singh Tiwari, PhD, , Mohd. Athar Kalam
Keywords
Related Articles
- EP ID EP116447
- DOI -
- Views 110
- Downloads 0
High Accuracy Detection and Tracking of Objects
One of the critical tasks in Computer Vision is Detection and Tracking of objects. But still now, the issues related to this are developing. For the automatic detection of moving objects, some of the monitoring s...
Heteroleach Protocol for Improvement of Stable Operation of Wireless Sensor Networks
Abstract : Wireless sensor networks(WSN) are very much sensitive to energy consumption by the sensor nodes. Many protocols have been designed for efficient use of energy of the nodes so that the lifetime of the network c...
Performance Analysis of Adaptive Approach for Congestion Control In Wireless Sensor Networks
WSN consists of hundreds / thousands of wireless nodes distributed within the geographical area. The wireless nodes gather information and supply towards the central node for further processing. There are different facto...
A Fast & Memory Efficient Technique for Mining Frequent Item Sets from a Data Set
Abstract: Frequent/Periodic item set mining is a extensively used data mining method for market based analysis,privacy preserving and it is also a heart favourite theme for the resarchers. A substantial work has be...
A Review on Content Based Medical Image Retrieval for Different Imaging Modalities, Organs and Diseases
This paper presents a review on Content based medical image retrieval for medical imaging modalities. Theevolution in the field of medical imaging system has lead industries to intellectualize a complete automated system...