ENHANCING RENTAL LISTINGS USING MACHINE LEARNING AND NLP

Abstract

India’s rental market has grown rapidly due to urbanization and population increases, particularly in cities like Mumbai, Delhi, and Bengaluru, where rental demand has surged by 20–30%. Despite this growth, most property listings lack data-driven insights into customer preferences. Traditionally, landlords relied on newspaper ads, brokers, and intuition to market properties and set prices, with little personalization or analytical feedback. This approach made it difficult to accurately gauge tenant interest and often resulted in inefficient, generic marketing. The proposed system addresses these limitations by integrating Natural Language Processing (NLP) and machine learning to analyze sentiments in property listings and viewer feedback. By processing descriptions and extracting key renter-attracting features, the system uses predictive models to classify sentiment from reviews, inquiries, and social media. This enables landlords to optimize pricing, tailor listings, and improve marketing strategies based on viewer preferences and past trends, ultimately enhancing discoverability, tenant engagement, and rental efficiency