© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 1756 House Price Prediction Using Machine Learning -A Survey Bharath P, Harshith V, Mohan Kumar G , Prema N S Department of Information Science and Engineering, Vidyavardhaka College of Engineering Mysuru, India ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - Methods for calculating the sale price of houses in cities remain a difficult and time-consuming task. The purpose of this article is to forecast the coherence of non-house prices. Using Machine Learning, which can intelligently optimize the optimum pipeline fit for a task or dataset, is a key technique to simplify the difficult design. Predicting the resale price of a house on a long-term temporary basis is vital, particularly for those who will be staying for a long time but not permanently. Forecasting house prices is an important aspect of real estate. The literature tries to extract relevant information from historical property market data. The price of real estate causes land price bubbles to expand, causing macroeconomic instability. The reasons that drive up real estate prices are important investigating so that the government may use them as a guide to help stabilize location, and various economic elements influencing at the time are all factors that influence the house selling price. Key Words: Machine learning, House price, Prediction, regression. 1. INTRODUCTION The value of a home is well known to be based on a wide range of factors. As a result, predicting the value of a home involves a unique set of issues. Houses are a need for society and rates vary depending on the amenities offered, such as size, area, location, and so on. Predicting the exact values of house pricing is a tricky process. This project is being suggested in order to better estimate property prices and provide more accurate results. This would be extremely beneficial to the people because house pricing is a problem that many individuals, rich and poor, are concerned about because one cannot gauge or predict the price of a property based on the location or amenities provided. Also, Professional appraisers are commonly used to anticipate house prices in the past. However, due to a huge interest from the people, house broker, buyer, or seller, an appraiser is likely to be biased. So as a result, an automated prediction system can be useful as an objective party source that is less biased. The price of a house is a time series. Various methods for estimating property prices have been offered. A house price prediction model seeks to figure out what elements influence price changes in a certain area. Clearly, the factors that influence housing prices are complicated and intertwined processes that typical statistical methodologies overlook. Despite the fact that the hedonic price model has gained widespread acceptance in recent years, it has been criticized for model assumptions and estimation, as well as for tackling nonlinear problems, global regression, and local clustering. To anticipate the variance in house prices, nonlinear machine learning and fuzzy logics were applied. In, a neural network was used to forecast property values. The Support Vector Machine was used with optimization techniques like the Generic Algorithm and Particle Swarm Optimization. Repeated Incremental Pruning to Produce Error Reduction, Nave Bayes, and Ada Boost were among the machine learning techniques studied in. In terms of estimating property price, the RIPPER algorithm surpasses other models, according to the study. Linear regression, decision trees, and nearest neighbor were used to estimate house prices. In addition, the study found that Nave Bayes was the most consistent classifier for unequal frequency distributions. Multiple linear regressions is a statistical approach for determining the relationship between numerous independent variables and the (dependent) target variable. The use of regression techniques to develop a model based on numerous criteria to forecast price is common. Predicting house prices is a difficult task. On the one hand, the factors that influence housing prices are complicated and vary nonlinearly, resulting in large forecast errors in standard models. On the other hand, the real estate market's daily data is massive and growing at a quick pace. The majority of recent research has focused on dismantling the distraction of house cost prediction. As a result of the analysis work done by various researchers all across the world, several theories have emerged. 2. Literature Survey Lu et.al proposed a hybrid prediction model; the study looked at the impact of land financing and household spending on real estate prices in 33 major Chinese cities. The implementation of Panel data validation of fixed- effects model regression findings our proposition After establishing control of the city's local people, the rate of growth, per capita GDP, and the number of students enrolled in regular classrooms are all things to think about. Institutions of higher education, gender ratio, and consumer pricing Higher education institutions, gender ratios, and consumer pricing urban population density, land finance, and urban development are all indices to look International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 09 Issue: 05 | May 2022 www.irjet.net p-ISSN: 2395-0072