Predicting protest flows in cyberspace based on data mining

Today, various groups in society use cyberspace to publish calls for illegal gatherings, followed by protests. Therefore, the main purpose of this study is to predict protest flows in cyberspace based on data mining.

The present study is applied in terms of purpose and descriptive-analytical in terms of method, in which the information data in documents, organizational documents as well as decentralized databases in police operational categories have been used. The statistical population of the study is all the information data of the invitations of illegal gatherings in cyberspace that were used for sampling from the data preprocessing method and in full. In order to predict and prevent crime, using RapidMiner software as an open source data mining tool in Java, first identify the main components of illegal aggregation calls in cyberspace, the relationships between the identified components in the data collection, and the data analysis. شد.

Among the various implemented models, the decision tree model with 91.39% accuracy provided the highest result for the desired data and prediction of protest flows, and its most important achievements were designing a model for predicting illegal gatherings and determining the effective components in gatherings. It is illegal.

The components of "ideology of the publisher of the calls, the number of visits to the calls, the sub-area of ​​the calls and the number of members of the sources and targets had the most to the least impact on the occurrence of illegal rallies and protests; In other words, the ideological attitude of the publisher of the calls plays a direct role in creating rallies, and the police can focus on it to predict the possibility of illegal rallies.