Modelling Machine Learning For Analysing Crime News
Surabhi Ghankutkar, Neelabh Sarkar, Pooja Gajbhiye, Sanyukta Yadav, Dhananjay R. Kalbande, Nida Bakereywala · 2019
Increasing crime rate and recurring crimes in India factors towards analyzing crime. Crime analysis is a way of observing the patterns of crime that happened recently and in past to get a better estimate of future crimes. Crime analysis is very important for the law enforcement and also helps crime detectives to find the criminals. Apart from this, common public also should be aware of the criminal activities so that they can take safety measures to protect their family and friends. Traditional methods of analyzing crime news involve reading the complete crime news and then analyzing it, which is very time consuming and vulnerable to human errors. Also many machine learning classifiers are used for classification of the old datasets of crime which will not be assistive enough to provide any report that analyzes crime news in real-time. The proposed system is a web-system which analyzes real-time crime data in the form of online news articles and gives a report of the news related to crime. The contents of the websites of newspapers are the input to the system. The site is crawled using a crawling program written in python language, and the data is stored in a temporary database. Using three different classifiers, the data is classified into crime-related data and non crime-related data. The final analysis wiil be displayed to the user in a tabular format with pie charts denoting crime and non-crime percentage.