Interactive Visual Exploratory Data Analysis (Eda) For Predicting House Prices In Metro City

Abstract

House price prediction is becoming increasingly popular, and research teams are increasingly doing similar studies utilizing deep learning or machine learning models. People are quite cautious when it comes to buying a new house based on market techniques and their budgets. Methods for exploratory data analysis can show the structure of data. Due to the abundance of available methodologies, picking ones that will function well together and facilitate data interpretation is a difficult issue. In this paper, we provide a well-fitting collection of tools for conducting a thorough exploratory investigation of a House price prediction dataset. The suggested technique consists of numerous steps, including robust data normalization, outlier handling, feature importance, missing value handling, and so on. By examining several criteria such as the house's area, square footage, number of bedrooms, and so on. The dataset from Kaggle was used in this study. Various graphical display techniques are used to undertake exploratory data analysis. This EDA is critical for using machine learning approaches and constructing models. According to the results of the investigation, the Kharghar location in Mumbai has the most available properties for sale. Also identified are the top ten Mumbai localities with the highest average property price for those who can afford luxury homes in Mumbai. EDA also aids in determining the relationship between the features that are significant in machine learning. Finally, the researcher used the folium library to create an interactive Mumbai map displaying the number of residences for sale by location.