Abstract:
Crime is a major problem faced today by society.
Crimes have affected the quality of life and economic growth
badly. We can identify the crime patterns and predict the crimes
by detecting and analyzing the historical data. However, some
crimes are unregistered and unsolved due to a lack of evidence.
Researchers used different sources to get crimes related data to
generate the prediction model. But, some crimes are
unregistered. In this paper, we used online news posts to detect
crimes. crimes. As the first step, we fetch the news posts using
predefined keywords relating to the crimes. Then, we proposed
the Long short-term memory (LSTM) approach to the
classification of crime types and non-crime related posts. Our
approach outperformed the existing approaches by obtaining
88.4% accuracy.
