A new efficient SIF-based FCIL (SIF-FCIL) mining algorithm in predicting the crime locations

摘要

In our innovative crime location forecast method, at the outset, the crime features are mined from the crime database and used for performing the adaptive mutation-based artificial bee colony (AMABC) algorithm, in which the database attributes and crime values are bunched together. Subsequently, the frequent closed itemsets lattice (FCIL) is built by the rules support factor values, and from this the frequent rules are extracted. In the course of the FCIL creation, the clustered attributes values are processed like a sliding window. In accordance with the frequent rules, the related crime locations are created. Thus, our proposed sliding with itemsets factor-based FCIL proposed technique is endowed with the superb skill of fruitfully forecasting the locations by means of AMABC and FCIL methods. In our innovative approach, we apply an UCI Machine Learning Repository-Communities and Crime Data Set for the offence investigation. The novel method is analysed and contrasted with the modern mining algorithms such as Apriori, Eclat and conservative FCIL.