This paper predicts the future house pricing for major cities in Pakistan including Islamabad, Rawalpindi, Lahore, Karachi, and Faisalabad. As stated by Al Jazeera, 2023 the inflation rate in Pakistan has been really high in the past couple of years and hence it has become significantly important for the residents to make important decisions regarding the purchase of houses and residential properties, therefore, this paper uses multiple Machine Learning models to predict the house pricing in the coming years based upon the prior available dataset. Major important factors and features such as the number of bedrooms, bathrooms, area, number of floors, garage, and so on have been taught while predicting future prices as most individuals base their decisions considering these factors when purchasing a house. The process that this study undergoes is based on prediction initiation which involves preparing the data, selection of the important features, and model training. The process of choosing the appropriate machine learning models played a crucial role as explained in the study. The significance of this study is not only limited to the buyers but in fact will enable the sellers, the policymakers, the government, and all other bodies linked with the real estate industry to have valuable information regarding the pricing so
that they can make well-informed decisions for their betterment. It is important to consider that our prediction accuracy is based mainly on factors such as the quality and quantity of available data, susceptible to evolving market dynamics, economic conditions, trends, and demands.

Briefly Discussion

Predicting house prices is a task, in today’s real estate industry involving individuals and businesses engaged in buying and selling properties. To ensure dependable property valuation information, Artificial Intelligence (AI) and machine learning models play a role. When it comes to predicting house prices we rely on sales data along with factors such as the number of bedrooms, bathrooms, size, location, and amenities of the houses. By leveraging these features machine learning models can estimate the selling price of a house enabling both buyers and sellers to make decisions. Accurate house rate predictions assist customers, sellers, sellers, and traders
make higher selections in the real estate market. However, we should maintain in mind that the accuracy of predictions relies upon the first-rate amount of statistics to be had. The actual property marketplace is dynamic and influenced by way of monetary situations, tendencies, and calls, which can affect the accuracy of predictions over the years. Ultimately, AI and gadget learning have emerged as precious tools for house charge prediction. By the use of ancient records and asset functions, system studying fashions can estimate house values, empowering stakeholders to make informed choices in the ever-changing actual property market.

Factors affecting House Pricing:

Multiple factors and features of a specific house can decide the price. Researchers in the past have proposed several considerable variables that play a vital role when it comes to influencing the overall housing price. These factors can be divided into three main categories that are, house factors, environmental factors, and transportation factors as mentioned by Kushan et al. Each of these key factors possesses an impact on the primary mechanism of pricing. Among all these factors, the most dominant type is the residential factor. The residential
factor is mainly linked with aspects like the living residence, utility, the number of rooms, and so on. These are factors that can alter the purchase decisions of the buyers in case they aren’t up to the mark and hence possess great significance. Extensive research of Basit et al., 2021 suggests that someone who chooses to live with family members would typically attach more importance to the essential features of the house, like the living area and the number of rooms, which have a significant impact on the overall living quality and experience in the house. In addition to intangible features, such as the view of the residence and usability, these aspects significantly impact housing prices by influencing buyers’ overall experience with the property and their willingness to pay. Some other factors upon which the purchase decision is based include building properties and floor factors. By building properties, we refer to the hardware and basic facilities that the
building has to offer, such as the elevator, generator, and garage. These can play an important role in the consumer’s purchase decision. For instance, people often form their decision based on whether the house had a garage or if the garage could fit all their cars and similar numerous aspects of the house. The number of vehicles being possessed by an average household is increasing with the passage of time, and therefore this factor holds a great bit of importance in the final purchase decision, as mentioned in Zhou et al., 2017 study.
Similarly, there are many other facilities that people like, such as the garden, the swimming pool, the backyard, the gallery view, and so on. All of these factors help determine the housing price since these things are all common among the facilities that modern-day houses have to offer, as mentioned by Basit et al., 2021. National Center for Biotechnology Information, n.d. found that similarly, people usually
prefer the number of floors that suit their daily convenience. A family with children has often been seen to choose multi-floor constructed houses so that get the appropriate privacy while living together. The environment that the house has to offer also is a major factor of
consideration. For instance, the surroundings that the house has to offer such as the view, the street, the locality, and the city are all indicators of the living quality and therefore have been given great importance by not only the marketers and the real estate agents but the
consumers as well. Based on the present literature, it was difficult to identify the most common variables/factors upon which the data had to be collected. Floors, Waterfront, view, sifts above and sqft basement, year built, year renovated, street, city, country, and zip code were identified as the key factors upon which data had to be collected. The literature also tells us that econometrics theory can be effectively used to develop a reasonable model to predict the single-house price in a target area or location. A regression model can be made that inculcates the effect of all the important variables as discussed earlier. The model that is developed using this approach is known as an econometric model is much more accurate at yielding the results when compared to the conventional models as stated by Hakimi et al., 2021.Therefore, after going through the techniques and methods present in the previous studies, my study is focused on one goal to establish the best linear unbiased estimator that enables us to predict future house prices based on necessary variables to educate amateur
house buyers in gaining a comprehensive understanding of the price determination process and the housing market so that they can form the best decisions for themselves. Bourassa et al.