Clustering Rental Houses in Bahrain Using K Modes Algorithm

Clustering is a key application in big data, essential for unsupervised learning. This study tackles the challenge of clustering a dataset with categorical housing attributes in Bahrain to uncover meaningful patterns and insights. Using the K-Modes algorithm, tailored for categorical data, the analysis includes features such as property type, location, number of bedrooms, and amenities. The performance is evaluated using the Silhouette score, which was 0.208, indicating moderate clustering effectiveness. The findings highlight distinct categories of residential properties in Bahrain, with Governorate being the feature with the highest differences among clusters. This research provides valuable insights into the Bahraini real estate market, aiding stakeholders in property development, urban planning, and housing policy formulation to make informed decisions and address market needs.