Tactical crime analysis using clustering on San Francisco crime data
- Pathum Thani, Thailand : Asian Institute of Technology, 2017
- 52 leaves : ill.
- Research studies project report ; no. IM-17-05 .
- Asian Institute of Technology. Research studies project report ; no. IM-17-05 .
A research submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Information Management, School of Engineering and Technology
Research studies project report (M. Eng.) - Asian Institute of Technology, 2017
In the digital era, police forces have access to quickly expanding sources of information. The enormous increase in the amount of available data has made the use of data mining techniques essential in nding important patterns. This research, on the data collected from kaggle competition,San Francisco Crime Classication, will take combinations of type of crime and time to determine locations where it is more likelytooccur. Thiswillhelpinplanningpreventivemeasures. Thoughthekagglecompetition expected the participants to determine the type of crime occurring given they have knowledge about the time and location, in this paper the outcome is slightly changed since many studies have been done on the before mentioned problem.To predict the outcome i.e. given the type of crime and time, at which location(s) its more likely to occur, in this study the technique used will be k-means clustering . ForperformingK-meansclusteringWekadataminingtoolwasused.Alsothedifferenttypesof crimes were analyzedon the basis ofdays in a week and graphsare providedto understand the relation between the rates of crime and the day it is happening on. In the results the location(s) of the most prominent crimes and time are given which will help the police forces to take preventive measures.
Data mining--Computer programs Cluster analysis--Data processing