The science of cohabitation: A study on roommate compatibility using machine learning

Authors

Shivanandaiah, Pradyumna

Issue Date

2024

Degree

MSc in Business Analytics

Publisher

Dublin Business School

Rights

Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.

Abstract

This thesis investigates the combination of psychological analysis and advanced data techniques to improve predictions of roommate compatibility. It emphasizes the importance of skilled data management and a deep understanding of psychology in creating effective predictive models. By integrating the Big Five personality traits—Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism—with machine learning methods such as Cosine Similarity and K-Means clustering, this research introduces a novel approach for evaluating compatibility in shared living spaces. The study employs a rigorous methodology that blends quantitative analysis with psychological insights, supported by a comprehensive dataset from Bustudymate’s student community and supplemented with secondary data from Openpsychometrics.org. This approach strengthens the foundation for the models used in predicting roommate compatibility. Findings reveal that the use of multi-dimensional data analysis and advanced algorithms significantly improves the accuracy of compatibility predictions, surpassing traditional matching methods. This thesis not only advances the application of machine learning in assessing social compatibility but also highlights its potential to make roommate matching a more objective and data-driven process. The contributions of this research extend across the fields of machine learning and social sciences, underscoring the critical role of quality data and accurate psychological assessment in developing effective predictive tools. The practical applications of this work are broad, providing improved resources for housing authorities and online platforms to ensure more harmonious communal living situations.