Paper Title
Optimizing Property Listings: A Machine Learning Framework for Real Estate Salability

Abstract
While machine learning has traditionally been applied to real estate price valuation, its use in predicting property salability remains largely unexplored. This paper introduces a novel machine learning framework specifically designed to forecast the likelihood of real estate properties being sold within a given timeframe. Unlike previous approaches that primarily focus on price estimation, our framework leverages diverse data inputs—such as property characteristics, market trends, and socio-economic factors—to optimize property listings for faster sales. By utilizing advanced machine learning algorithms, the proposed model aims to provide more accurate predictions of property salability, addressing a significant gap in current real estate analytics. Initial results suggest that this data-driven approach can reduce time on market (TOM) and enhance closing rates by optimizing key factors like location, pricing, and listing strategies. This novel framework is expected to offer substantial benefits to real estate agents, developers, and property owners by enabling informed decision-making to align listings with market demand, thereby maximizing sales efficiency. Keywords - Real estate success, salability, machine learning, XGBoost, LightGBM, gated communities, time on market, Egypt, predictive models