Big Data-Driven AI Analysis of Real Estate Valuations in the Residential Sector of Tashkent City by Advanced Data Analytics Techniques in Terms Market Dynamics
Abstract
This paper aims to analyze the trends in real estate in Tashkent using data from a database of property listings at Uybor.uz obtained from January 2015 to September 2023. For real estate in the metro area, the missing values are estimated using the mean of the IQR method and the outliers in numerical variables are cleaned using standardization. These are the elements that make up a comprehensive description and are extended to the distributions as well as the inter-variable relationships where the Pearson correlation coefficient for Envelope linear Outlier relationships Detection is and also supported given. by The Mahalanobis method Distance, for which detecting reveals outliers that is there they are Elliptic 186 outliers in the data (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{2. 5 1} \boldsymbol{\%}$</tex>). In LASSO regression modeling of data some overfitting is prevented by some of the coefficients being reduced to zero and some variables being chosen. It has been seen that there is a very high significance of the linear relationship of the size of the property with the price of the property which means the implications of the data driven market insights. These methodologies provide a strong basis for real estate appraisal and market assessment so that the stakeholders can get to know the real situation of the market and the evaluation models can be improved.