Research Article
Classification of Criminal Crimes From Data Twitter Using Class Association Rules Mining
@INPROCEEDINGS{10.4108/eai.20-1-2018.2281925, author={Husna Gemasih and Rayuwati Rayuwati and Azhari SN and Mursalin Mursalin}, title={Classification of Criminal Crimes From Data Twitter Using Class Association Rules Mining}, proceedings={Proceedings of the 1st Workshop on Multidisciplinary and Its Applications Part 1, WMA-01 2018, 19-20 January 2018, Aceh, Indonesia}, publisher={EAI}, proceedings_a={WMA-1}, year={2019}, month={9}, keywords={twitter data class association rules laplace accuracy}, doi={10.4108/eai.20-1-2018.2281925} }- Husna Gemasih
Rayuwati Rayuwati
Azhari SN
Mursalin Mursalin
Year: 2019
Classification of Criminal Crimes From Data Twitter Using Class Association Rules Mining
WMA-1
EAI
DOI: 10.4108/eai.20-1-2018.2281925
Abstract
The police obtain criminal crime data from the field based on reports from a person or group, from the data the police can evaluate the crimes that occurred. The police have no reports of crimes from other parties such as from social media. One such social networking media is Twitter. The information conveyed by Twitter users in a tweet usually contains something related to himself or his environment, including the occurrence of a crime. Such information will serve as data for classification as well as to know the trends of criminal crime. This research uses classification data mining technique that is Class Association Rules (CARs). CARs will find all frequent ruleitems through a series of stages and build rules using frequent ruleitems obtained, then rules will be obtained. The resulting rule will be evaluated to determine the strength of the rule using the Laplace Accuracy equation, which will produce the best rule. These rules will serve as models for the new data classification. The result of accuracy test of this method by using 100 test data is 96%.