Tenant harassment is a common occurrence affecting many people around the world. However, many less privileged people have had un-told hardships from greedy and gain-loving landlords or agents who use a variety of tactic to make life uneasy for tenants. This research, therefore, develops a machine learning algorithm to predict the possibility of a tenant being harassed. Online dataset about tenant harassment having eleven attributes was downloaded. The research extracted useful features from the dataset using lasso algorithm, Naive Bayes and Logistic Regression was used for predicting whether there is harassment or not. The result shows that logistic regression performs slightly better with an accuracy of 95.4% while naive Bayes has an accuracy of 94.7%.

Authors: Otasowie Owolafe, Samuel A. Adegbite

Published in: International Conference for Internet Technology and Secured Transactions (ICITST-2021)

  • Date of Conference: 7-9 December 2021
  • DOI: 10.20533/ICITST.2021.0023
  • ISBN: 978-1-913572-39-6
  • Conference Location: Virtual (London, UK)