MAIL MESSAGE FILTERING BASED ON ARTIFICIAL INTELLIGENCE
Authors
Fedir Prokhnytskyi
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Rokovyi Oleksandr
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
Abstract
Abstract: The purpose of the work is to analyze the effectiveness of the email filtering module. The research uses a dataset from the Kaggle platform that has been processed and supplemented with additional messages, as well as classifiers based on two different models: a naive Bayesian classifier; support vector machines method. The effectiveness of each of the approaches based on the same sample was analyzed using model metrics: precision and recall. Each model was further tested on the test data set.
Keywords: artificial intelligence, machine learning, classifier, neural network.
