House Price Prediction of Real Time Data (DHA Defence) Karachi Using Machine Learning

Abstract

Pakistan’s real estate market has a large impact in GDP growth. Investment in real estate sector in Pakistan is encumbered with lucrative opportunities. The market demand for housing is ever increasing year by year. House sales prices keep on changing and increasing frequently, so there is a need for a system to forecast house sales prices in the future. Several factors that influence house sales price includes; location, physical attributes, number of bedrooms as well as several other economic factors. One of the main motivation of choosing Karachi for the house prediction is that Karachi is capital of Sindh and it has significant importance in country's economic as it is the major commercial and industrial center of Sindh. It is one of the main contribution of the work is that through this the house prediction model based on DHA Karachi data is developed and as per best of our knowledge till today there is no prediction of housing for the country’s important has been developed. has This research paper mainly focuses on real time Defense Housing Authority (DHA) Karachi data, applying different regression algorithms like Decision tree, Random forest and linear regression to find the sales price prediction of the house and compare the performance of these models. Random Forest algorithm gives 98% of accuracy. The proposed work will be very much helpful for the common people, real-estate people, investors and builders to inform them about making decision of selling or buying at Defense Housing Authority (DHA) Karachi.