KNN CLASSIFIER AND NAÏVE BAYSE CLASSIFIER FOR CRIME PREDICTION IN SAN FRANCISCO CONTEXT

 

KNN CLASSIFIER AND NAÏVE BAYSE CLASSIFIER FOR CRIME PREDICTION IN SAN FRANCISCO CONTEXT

Noora Abdulrahman and Wala Abedalkhader

Department of Engineering Systems and Management, Masdar Institute of Science and Technology, Abu Dhabi, the United Arab Emirates

ABSTRACT

In this paper we propose an approach for crime prediction and classification using data mining for San Francisco. The approach is comparing two types of classifications: the K-NN classifier and the Naïve Bayes classifier. In the K-NN classifier, two different techniques were performed uniform and inverse. While in the Naïve Bayes, Gaussian, Bernoulli, and Multinomial techniques were tested. Validation and cross validation were used to test the result of each technique. The experimental results show that we can obtain a higher classification accuracy by using multinomial Naïve Bayes using cross validation.

KEYWORDS

Classification, K-NN Classifier, Naïve Bayes, Data Mining, Gaussian, Bernoulli, Multinomial, Uniform, Inverse & Python

Full Text: https://aircconline.com/ijdms/V9N4/9417ijdms01.pdf

Volume Link: https://airccse.org/journal/ijdms/current2017.html