HOUSE PRICE PREDICTION USING LINEAR REGRESSION IN ML

Bhumireddy Vishnuvardhan reddy, S.R Veera Siva Reddy

Student, Department of Computer Science & Engineering(AI), Vivekananda

The purpose of the paper is to predict the market value of the property being sold. This program helps to find the
startingprice of a location based on location variables. Similarly, consider a situation in which a person needs to sell
a house. Byusing a real estate pricing system, the seller will be able to determine what features he can add to the
house so that the house can be sold at a higher price. Therefore, in both cases, we can be sure that the home price is
good for both the buyer and the seller. Housing prices goup every year, so there is a needfor a real estate forecasting
system. Estimating the price of ahouse can help a developer determine the selling price of a house and can help
clients set a reasonable time to buy a home.

Keywords: linear regression model, Python, Machine Learning.

Supervised Learning, Data science, House Price

Machine learning has played a major role since years ago in selecting an image, spam, common speech command.
Machine learning also provides better customer service and safer car systems. The home-market market stands out
among the most focused on prices and is constantly evolving. It is one of the main areas in which to apply machine
learning ideas on how to improve and predict costs with high accuracy. A single heuristic database is often used to
analyse the decline in house prices in the Boston city housing database.Previous analysis has found that house
prices in that database depend largely on their size and location. Until recently, basic algorithms such as linear
regression could reach 0.113 predictive errors using both internal building features (living space, number of rooms,
etc.) and additional spatial features (demographic image features such as revenue rating, population. congestion,
etc.). Modelling uses machine learning algorithms, in which the machine learns data and uses it to predict new data.
As we know, the proposed model for accurately predicting future outcomes has economic applications, business,
banking, health care sector, e-

commerce, entertainment, sports etc. One such method used to calculate real estate is based on anumber of factors.
In large metropolitan areas such as Bengaluru, a potential buyerconsiders a few factors such as location, size,
proximity to parks, schools, hospitals, power stations, and most importantly thevalue of the house.

Machine learning: Machine learning is a data-based system learning process. Machine learning is part of data
science where we use machine learning algorithms to process data.

Supervised reading strategy It is a hypothetical model used for activities where it involves predicting one value
using other values in a data set. Supervised reading will have pre- defined labels. Separates an object based on
parameters in one of the predefined labels.

Supervised learning method Descriptive model is used for activities that can benefit from the understanding gained
bysummarizing data in new and exciting ways. There are no predefined labels in unattended reading mode. In the
real estate market, data is a very important source of analysis and forecasting. It is always beneficial to know about
the speculation of
Literature Review

A browser-based application intended for real estate. The proposed system is a standard system that can be accessed
in different locations. System overcomes with all the different barriers we have in the existing system and comes up
with a solution. The main purpose of the system is to predict the price of a home. The proposed system uses a
machine learning method to predict the price.

We use a supervised reading method to predict. From downloaded online training data sets, data collected from
sources such as Kaggle and global data websites. we use many parameters to predict theprice of a house. The
proposed system is an application based on an Internet browser. The real-time app where we can access the app is in
a different location. The main purpose of this program is to predict the price of a home. The proposed system uses
parameters such as house size, balcony, bathroom number, location

and other parameters to predict house price. The systemuses a line regression model to predict the price. We use
effective price predictors. The system helps wealth to make quick decisions.

System Architecture
Screenshot: 1. Showing different imported modules

Experimental Setup:

Solution to the problem is how toback off the line. This process modelled the relationship between target variables
and independent variables (predictions). It equates the line model with the coefficients in the data to reduce the
residual amount of squares between the target variation in the data, and the predicted values by line measurement.
This data set has many features, including:

• ••Estimated income in the area ofthe house

Household items can be divided into several types. The most influential type are the living space, which includes
the living space, usability, and number of rooms. When people think of buying a house for a living, the above
factors are a key factor in the quality of life. Buyers with family members can often attach additional importance to
an important aspect of the home, such as living space and numberof rooms, which have a significant impact on the
overall quality of life and the knowledge of the house. In addition, intangibles, such as residential and usable views,
also have a significant impact on housing prices, both affecting buyers' knowledge of the house and
theirwillingness to pay.

The linear regression performs the function to predict the variable amount of response (dependent) (y) based on a
given(independent) descriptive variance (x). Therefore, this regression method detects the linear relationship
between x (input) and y (output).

Keep in mind that the X and Y volatility values are training datasets to represent the line regression model.

Consider a real estate companywith databases containing real estate numbers for a particular region. The price of a
Methodology

The block diagram given below is the summary of the methodology followed in the paper.

• How to do it
• The block diagram provided below is a summary of the processfollowed on paper.
• Illustration: block drawing

2. 1.Data collection and analysis:

My process can be divided into several stages. The first stage is the data collection phase where Icollected the data
online. This will be used to train the machine learning model. The data collected in this category is raw and
unstructured data. Accordingto the data.The price column in the database is a dependent variable and some columns
are independent variables (also called features).

• This data set has many features, including:

Screenshot 3

Screenshot 4 : data clean

Fig: heat map

https://www.researchgate.ne t/publication/349098061_Sim ple_Linear_Regression_Hou

nload

https://www.ijert.org/compari

son-of-machine-learning- algorithms-for-house-price- prediction-using-real-time- data

implilearn.com/tutorials/pyth on- tutorial/data-visualization-in- python

S.Abhishek.:Ridge regression vs LassoHow towork these two

popularmethods of ML regression.AnalyticsIndia

Magazine,2018.

H.L. Harter, Small Square Method and other ves-Part II.International

Static

Review.1972,43 (2)

, pp. 125-190
Future Enhancements
We can add an engineer where buyers can buy a house online and can complete the process.

We can add a query module where guests can submit questions.