Write about the Assignment on Business Intelligence for Crime.
The social problem that adversely affects the quality of life and dampens the economic growth of any society is generally termed as Crime. It is the unwanted action of individual people or a gang of people that takes place at certain time of the day and at certain location that distorts the social balance of any region. In present scenario, prediction of crime and its pattern has been a matter of concern for the department of criminology of different country.
One of the contemporary research on this topic highlights that researcher tries to work on the existing literature and elucidate the importance of news and other media channels in influencing the crime pattern and rate (Mookiah, Elberle and Siraj 2015). Another research based on Merseyside (Milo et al. 2013), UK highlights the existence of correlation between the increased crime rate and spatial location of its occurrence. Research carried out by Dr. Vijayakumar, Dr. Karthick & Praksh (2013), used GIS and clustering simulation in identifying the pattern and thereby forecasting the possibility of criminal activity in Tamil Nadu, India.
According to Almanie, Mirza and Lor the unethical actions takes place based on certain patterns. The authors compared their work and stated that in spite of the fact that there has been many research carried on in this genre but none of the contemporary literature considered the multi correlation between time, location and crime type.
The main aim of the work by Almanie and his fellow researchers is to explore the existing available data set in finding out the pattern of crime and identifying the spatial location of these criminal activities. After finding out the pattern, the study tries to generate a model that might help in predicting the future tile and location of upcoming criminal activity.
The chosen paper mainly tries to identify the types of crime and predict the future occurrence of the same. Data set have been collected for the region of Los Angeles and Denver. Two tools namely Naive Bayesian classifier and Decision Tree classifier have been used to predict the crime type in this paper. In the contemporary works map application has been used in predicting the crime type. The researchers have collected the data and initially categorised them into 14 different crime types. In order to obtain a frequently occurred pattern, the types have been transformed into 6 types of crime occurring in these two regions. These six types of crime are assault, taking drugs alcohol, other crimes, public disorder, theft and white collar crimes (Almanie, Mirza and Lor 2015).
After putting the data in tabular form and constructing the graph it has been observed that the problem of theft is the mostly occurred incident in both Los Angeles and Denver. The prediction also showed that occurrence of theft has been evenly distributed throughout the different days in the week. It has been seen that Naive Bayesian model predicted the crime type more accurately than the Decision tree model. The calculation has been done on the basis of Bayes theorem whereby the independent effects between the different attributes have been measured by Multinomial Naive Bayes. In doing so it used 4 variables namely the month of occurrence, day when it occurred, the time and spatial location of the occurrence of crime. In Denver, the model showed 51% accuracy whereas in Los Angeles it was 54% accurate (Almanie, Mirza and Lor 2015).
The different research on criminology has shown different patterns in crime rate, crime type, time and month of occurrence and so on. In this particular given paper it has been observed that three variables namely location, time and type of crime. However, the study showed that the crime had occurred in region where more youths reside than on regions where more middle aged people and old people resides. Again in region where the demography is bias towards male dwellers are prone to experience higher level of criminal activity than the region which is dominated by female residents. Hence for further study few more variables can be included to study the pattern of criminal activities of any region. Research may be conducted by considering income as a variable (Enamorado et al. 2016). In addition, the average educational qualification of the people in particular regions can also be taken as a variable along with the demographic pattern of the same region. This might yield some more interesting, unknown fact about the occurrence of crime and help the society in establishing a better law enforcement thereby safeguarding the people of that particular society.
Almanie, T., Mirza, R. and Lor, E., 2015. Crime Prediction Based On Crime Types And Using Spatial And Temporal Criminal Hotspots. arXiv preprint arXiv:1508.02050.
Enamorado, T., López-Calva, L.F., Rodríguez-Castelán, C. and Winkler, H., 2016. Income inequality and violent crime: Evidence from Mexico's drug war. Journal of Development Economics, 120, pp.128-143.
Milo, M.W., Richards, S.C., Saraf, P., Malambo, L. and Bhattacharjee, T., Crime Hotspot Tracking and Geospatial Analysis in Merseyside, UK.
Mookiah, L., Eberle, W. and Siraj, A., 2015, April. Survey of Crime Analysis and Prediction. In FLAIRS Conference (pp. 440-443).
VijayaKumar, M., Karthick, S. and Prakash, N., 2013. The day-to-day crime forecasting analysis of using spatial-temporal clustering simulation. International Journal of Scientific & Engineering Research, 4(1), pp.1-6.