Title of Publication: Ward’s Clustering Algorithm for Detecting Higher and Lower Areas of Crime Concentrations
Author(s): S.U. Gulumbe, H.G. Dikko, Y. Bello
Year of Publication: 2012
In this paper, the Ward algorithm method was used on two different distance methods to classify KatsinaState local government areas into homogenous clusters of crimes. The distance methods were Euclidean and Chi-Squared distances. Using the algorithm on the Euclidean distances between the original observations, we determine the most common crimes in the state by averaging the original observations within the clusters obtained. Using the algorithm on the Euclidean distances between the standardized observations, we determine the clusters of higher and lower crime concentrations by averaging the standardized observations within the clusters obtained. By using the algorithm on the chi-squared distance between the rows observations, we determine the popular crimes in each of the clusters by averaging the relative frequencies within the clusters obtained. Geographic Information System has also been used to delineate areas of higher and lower crime concentrations. The result of the analysis has shown that Katsina and Funtua local government areas have the highest crime concentrations in the state.
Key words: Geographical Information System, Ward’s Algorithm, Euclidean distance, Chi- Squared distance.
Need complete download? Login to download