Prediction and Comparison of Crime Rate using Advanced Machine Learning Algorithms and Bio inspired Computing Techniques
Crimes are the increasing danger to the mankind and India is facing one of the vast effects among them. There are numerous crimes that happens normal interim of time. Since Crimes are expanding, there is a need to illuminate the cases and processes in a lot quicker way. The Crime exercises have been expanded at a quicker rate and it is the obligation of crime analysis department to control and decrease the crimes exercises. There is a need of innovation through which the case explaining could be quicker. Procedure and techniques of crime analysis is the process of recognizing and sorting conclusions communicated in a piece of content, particularly with the end goal to decide if the item is positive, negative or unbiased along with comparing with Unsupervised and supervised processes of Machine learning to get better accuracy, precision and recall percentages along with graphs and charts to represent current state of crime activities in the locality mentioned in data. Paper explains “how the fetching data from the source”, pre-processing, sorting is done to implement and classify. The whole process gives a brief of the advanced subjects of Machine Learning Algorithms that describes the subject of Crime analysis and comparison with ensemble of classification techniques and unsupervised algorithms. The supervised and unsupervised algorithms will be compared with each other on the basis of accuracy. This paper explains the use of Genetic algorithm for the selection of features that when used for a particular classifier will result in highest possible accuracy. The more the accuracy come across according the dataset given the more the result will be up to the mark. The comparison and the use of Genetic algorithm provides the best accuracy and list out the features that have higher co-relation with the crime type.
Keywords: Naïve Bayes, SVM, Gradient boosting, Adaboost, Random Forest, Crime, Genetic, Ensemble