Price Prediction for Indian Cities Housing Property using KNN Algorithm

TABLE OF CONTENTS

1. INTRODUCTION………………………………………………………………….2

1.1. Aim of the Project ……..……………………….………………………...…………..2

1.2. Overview of the Project…………………………………………….……................... 2

2. SYSTEM ANALYSIS AND DESIGN ………………………………………………..3-4

2.1. System Analysis………………………………………………………………………3

2.2. Architecture Overview…………………………………………………………..........3
2.3. Module Description………………………………………………………………......4
2.4. Data Flow Explanation………………………………………………………….........4

3. REQUIRMENTS ………………………………………………………………...........5-6

3.1. Software Requirements………………………………………………………….…....5

3.2. Software Description………………………………………………………..………...5
3.3. Hardware Requirements………………………………….…………………..……….6

4. IMPLEMENTATION ..................................................................................................7-11

4.1. Data Collection and File Upload Functionality ……………………………………...7

4.2. Price Analysis Methodologies …………………………………….…………….....7-8
4.3. Comparative Analysis Tools …………………………………………………............8
4.4. Frontend Implementations…………………………………………………….…..8-10
4.5. Backend Implementations……………………………………………………......10-11

5. RESULT ANALYSIS ……………………………………………………………....12-14

5.1. Data Visualization …………………………………………………………....….…12

5.2. Screenshots……………………………………………………………………....12-14

6. CONCLUSION AND FUTURE WORK………………………………………...........15

6.1. Conclusion………………………………………………………………….…….....15
6.2. Future Enhancement…………………………………………………………….......15

7. REFERENCES…………………………………………………………………………16

Dept. of CS&E, MIT Mysore. 1

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 1
INTRODUCTION
The housing market in Indian cities is a critical component of the country's economy,
influencing various sectors and impacting millions of lives. With rapid urbanization, economic
growth, and evolving government policies, the dynamics of housing property prices in Indian
cities have become increasingly complex and significant. This project, titled " Price Prediction
for Indian Cities Housing Property using KNN Algorithm," seeks to delve into the multifaceted
nature of housing property prices across major Indian cities. Predicting housing property prices
in India is a multifaceted task that involves analyzing a variety of economic, demographic, and
market-specific factors. Economic indicators such as GDP growth, interest rates, and inflation
play crucial roles, as they directly impact the purchasing power and affordability for potential
buyers. Additionally, demographic trends like population growth and urbanization
significantly influence housing demand, with more people moving to cities leading to increased
demand for urban housing.

1.1 Aim of the Project

The aim of this project on " Price Prediction for Indian Cities Housing Property using
KNN Algorithm " is to analyse historical and current trends in housing property prices across
various Indian cities. It seeks to identify and evaluate the factors influencing these prices, such
as economic conditions, government policies, and urban development. By developing
predictive models, the project aims to forecast future housing property prices based on
identified trends and factors. Additionally, it will conduct a comparative analysis of property
prices between different cities and regions within India.

1.2 Overview of the Project

The project titled " Price Prediction for Indian Cities Housing Property using KNN
Algorithm " aims to provide a comprehensive analysis of the housing market across major
Indian cities. This includes examining historical data to identify trends in property prices and
understanding the economic, social, and policy factors that influence these trends. The project
will employ statistical and machine learning models to predict future property prices, offering
valuable insights for potential investors and stakeholders. The project aims to evaluate the
effectiveness of the KNN algorithm in accurately predicting housing prices based on various
factors and city-specific dynamics in India.

Dept. of CS&E, MIT Mysore. 2

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 2
SYSTEM ANALYSIS AND DESIGN
2.1 System Analysis

 Requirements Analysis: Define functional and non-functional needs, focusing on data

accuracy, scalability, and user interface requirements.
 Data Sources: Integrate data from real estate listings, government reports, economic
indicators, and demographic information.
 System Architecture: Use relational or NoSQL databases for storage, and ETL tools
for data cleaning and preprocessing.
 Analytics Engine: Implement statistical analysis and machine learning models for
trend analysis and price prediction.
 Visualization: Employ visualization tools for creating dashboards and generating
comprehensive reports.
 Security and Privacy: Ensure data protection through encryption, access control, and
secure data transmission.

2.2 Architecture Overview

 Data Sources: Collect data from real estate listings, government reports, economic
indicators, demographic information, infrastructure projects, and market surveys.
 Data Collection: Use APIs,
web scraping, and direct
database connections to gather
and integrate data from various
sources.
 Data Storage: Store data in
scalable relational or NoSQL
databases, ensuring efficient
management of structured and
unstructured data.
 Data Processing: Apply ETL
processes to clean and preprocess data, including handling duplicates and standardizing
formats.

Dept. of CS&E, MIT Mysore. 3

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

 Analytics and Modelling: Utilize statistical analysis and machine learning models to
analyse trends and predict future property prices.
 Visualization and Reporting: Create interactive dashboards and reports using
visualization tools to present insights and trends effectively.

2.3 Module Description

 Data Collection Module: Gathers data from real estate listings, government reports,
economic indicators, and market surveys using APIs, web scraping, and database
connections.
 Data Cleaning and Preprocessing Module: Handles missing values, removes
duplicates, and standardizes data formats to ensure data quality and readiness for
analysis.
 Data Storage Module: Stores structured and unstructured data in scalable relational or
NoSQL databases, ensuring efficient data management and access.
 Analytics and Modeling Module: Applies statistical analysis and machine learning
algorithms to identify trends, patterns, and make predictions about property prices.
 Visualization and Reporting Module: Creates interactive dashboards and reports to
present insights and trends in property prices effectively.
 User Interface Module: Provides a web-based or desktop application for users to
interact with the system, view reports, and access visualizations.

2.4 Data Flow Explanation

 Data Collection: Gather data from various sources such as real estate websites,
government databases, and market surveys using APIs, web scraping, and direct
connections.
 Data Ingestion: Store collected data in a centralized database system for easy
management and retrieval.
 Data Cleaning and Preprocessing: Clean and standardize the data by removing
duplicates, handling missing values, and correcting inconsistencies.
 Data Analysis and Modelling: Apply statistical analysis and machine learning
algorithms to identify trends, patterns, and make predictions about property prices.
 Data Visualization: Create interactive dashboards and reports to visualize trends and
insights from the analysed data

Dept. of CS&E, MIT Mysore. 4

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 3
REQUIREMENTS
3.1 Software Requirements

 Data Collection Tools: Web scraping libraries (e.g., BeautifulSoup, Scrapy) and APIs
for retrieving data from real estate platforms and government sources.
 Database Management System: Relational databases (e.g., PostgreSQL, MySQL) for
structured data, and NoSQL databases (e.g., MongoDB) for unstructured data.
 Data Processing and ETL: ETL tools (e.g., Apache NiFi, Talend) and data cleaning
libraries (e.g., Pandas) for data transformation and preparation.
 Analytics and Modeling: Statistical and machine learning libraries (e.g., Scikit-learn,
TensorFlow) for analyzing trends and building predictive models.
 Visualization Tools: Dashboard and visualization software (e.g., Tableau, Power BI)
for creating interactive reports and data visualizations.
 User Interface: Web development frameworks (e.g., React, Angular) and backend
frameworks (e.g., Django, Flask) for developing user interfaces and server-side logic.

3.2 Software Description

 Data Collection Tools: Web scraping libraries (e.g., BeautifulSoup, Scrapy) and APIs
for gathering data from real estate platforms and government sources.
 Database Management System: Relational databases (e.g., PostgreSQL, MySQL) for
structured data, and NoSQL databases (e.g., MongoDB) for unstructured data.
 Data Processing and ETL Tools: ETL tools (e.g., Apache NiFi, Talend) for
transforming data, and data cleaning libraries (e.g., Pandas) for preprocessing tasks.
 Analytics and Modelling Software: Statistical and machine learning libraries (e.g.,
Scikit-learn, TensorFlow) for analysing trends and building predictive models.
 Visualization Tools: Dashboard and visualization software (e.g., Tableau, Power BI)
for creating interactive reports and visualizing data insights.
 User Interface: Web development frameworks (e.g., React, Angular) and backend
frameworks (e.g., Django, Flask) for developing user interfaces and managing server-
side logic.

Dept. of CS&E, MIT Mysore. 5

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

3.3 Hardware Requirements

 Server Hardware: High-performance servers with multi-core processors, at least

16GB of RAM, and SSD storage for data processing and analytics.
 Database Servers: Dedicated servers with powerful CPUs, 32GB or more of RAM,
and reliable storage solutions for managing database operations.
 Workstations: Development machines with modern processors, a minimum of 8GB
RAM, and high-resolution displays for development and analysis tasks.
 Network Infrastructure: High-speed routers, switches, and network cables to ensure
fast and secure data transfer between system components and users.
 Backup and Storage Solutions: Backup systems and additional storage devices (e.g.,
NAS) for data protection and redundancy.
 Cloud Infrastructure: Cloud services (e.g., AWS, Azure) for scalable computing
resources, storage, and processing power as needed.

Dept. of CS&E, MIT Mysore. 6

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 4
IMPLEMENTATION
4.1 Data Collection and File Upload Functionality

1. Data Collection
 Automated Retrieval: Use web scraping tools and APIs to collect property price
data from websites and databases.
 Database Integration: Connect to existing databases to fetch historical data and
update the project’s database.
 Scheduled Fetching: Set up automated tasks to periodically retrieve updated data.
 Data Verification: Implement checks to ensure data accuracy and completeness.
2. File Upload Functionality
 User Interface: Provide a web form for users to upload data files (e.g., CSV,
Excel).
 Server-Side Processing: Handle and validate file uploads on the server.
 Data Parsing: Use libraries to parse and integrate file data into the database.
 Error Handling: Provide feedback on file upload issues.
 Data Storage: Store the processed data in the database.
 Security: Implement access controls to secure file uploads and data access.

4.2 Price Analysis Methodologies

The price analysis methodology for the " Price Prediction for Indian Cities Housing
Property using KNN Algorithm " project starts with data preparation, which involves cleaning
the dataset by addressing missing values, removing duplicates, and standardizing formats,
followed by integrating data from multiple sources into a unified dataset. During exploratory
data analysis (EDA), descriptive statistics and visualizations such as histograms and scatter
plots are used to understand the distribution and identify patterns in property prices. Feature
engineering then involves creating and selecting relevant attributes, and encoding categorical
variables if necessary. Statistical analysis is performed to explore correlations between
property prices and various features, and regression analysis is applied to quantify the impact
of different factors. Predictive modeling involves selecting and training machine learning
models, such as decision trees or gradient boosting, to forecast property prices.

Dept. of CS&E, MIT Mysore. 7

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

4.3 Comparative Analysis Tools

The comparative analysis tool for the " Price Prediction for Indian Cities Housing
Property using KNN Algorithm " project is designed to evaluate and compare property prices
across various cities effectively. It integrates data from multiple cities, ensuring that property
features and pricing attributes are consistently formatted for accurate comparisons. The tool
provides interactive visualizations, such as heatmaps, bar charts, and scatter plots, to visually
represent and compare property prices and trends.

It supports benchmarking by allowing users to set reference points for comparison, and
employs statistical methods like ANOVA or t-tests to identify significant differences and
influencing factors. Additionally, the tool tracks price trends over time across cities to reveal
growth patterns and market dynamics. Its user-friendly interface enables users to easily select
cities, specify parameters, and generate comprehensive comparative reports and visualizations,
facilitating informed decision-making and strategic insights.

4.4 Frontend Implementations

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Property Search</title>
<link rel="stylesheet" href="style.css">
</head>
<body>
<header>
<div class="logo">Real Estate Co.</div>
<!-- <nav>
<ul>
<li><a href="#">Home</a></li>
<li><a href="#">About</a></li>
<li><a href="#">Services</a></li>

Dept. of CS&E, MIT Mysore. 8

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

<li><a href="#">Contact</a></li>
</ul>
</nav> -->
</header>
<main class="main-content">
<div class="container">
<h1>Property Search</h1>
<form action="/search" method="POST" class="search-form">
<div class="form-group">
<label for="city">City:</label>
<input type="text" id="city" name="city" required>
</div>
<div class="form-group">
<label for="min_price">Minimum Price:</label>
<input type="number" id="min_price" name="min_price" required>
</div>
<div class="form-group">
<label for="max_price">Maximum Price:</label>
<input type="number" id="max_price" name="max_price" required>
</div>
<div class="form-group">
<label for="min_area">Minimum Area (sq):</label>
<input type="number" id="min_area" name="min_area" required>
</div>
<div class="form-group">
<label for="max_area">Maximum Area (sq):</label>
<input type="number" id="max_area" name="max_area" required>
</div>
<div class="form-group">
<button type="submit">Search</button>

Dept. of CS&E, MIT Mysore. 9

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

</div>
</form>
</div>
</main>
<footer>
<p>&copy; 2024 Real Estate Co. All rights reserved.</p>
</footer>
</body>
</html>

4.5 Backend Implementations

import express from 'express';

import path from 'path';
import bodyParser from 'body-parser';
import { spawn } from 'child_process';
import XLSX from 'xlsx';
const app = express();
const port = 3000;
app.set('view engine', 'ejs');
app.set('views', path.join('views'));
app.use(bodyParser.json());
app.use(bodyParser.urlencoded({ extended: true }));
// Serve static files (e.g., CSS, images)
app.use(express.static('public'));
// Routes
app.get('/', (req, res) => {
res.render('index');
});
app.post('/search', async (req, res) => {
const { city, min_price, max_price, min_area, max_area } = req.body;
// Process the input data as needed (e.g., filter properties)
// Example: Log the received data
let a = [city, max_price, min_price, max_area, min_area];

Dept. of CS&E, MIT Mysore. 10

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

// You can render another view or perform other actions here

await runPythonScript(a);
const workbook = XLSX.readFile('output.xlsx');
const sheetName = workbook.SheetNames[0];
const sheet = workbook.Sheets[sheetName];
// Convert the sheet to JSON object
const data = XLSX.utils.sheet_to_json(sheet);
// Display the JSON object
console.log(typeof(data));
res.render('display', { propertiesData: data });
});
function runPythonScript(a) {
return new Promise((resolve, reject) => {
const pythonProcess = spawn('py', ['./index.py', ...a]);
pythonProcess.stdout.on('data', (data) => {
resolve(data.toString());
});
pythonProcess.on('error', (error) => {
reject(error);
});
pythonProcess.on('close', (code) => {
if (code !== 0) {
reject(Python process exited with code ${code});
}
});
});
}
// Start server
app.listen(port, () => {
console.log(Server is running on http://localhost:${port});
});

Dept. of CS&E, MIT Mysore. 11

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 5
RESULT ANALYSIS
5.1 Data Visualization

In the " Price Prediction for Indian Cities Housing Property using KNN Algorithm "
project, data visualization is a pivotal component that transforms complex data into accessible
and actionable insights. The project utilizes a range of visualization techniques to provide a
comprehensive understanding of property pricing across different cities. Central to this is an
interactive dashboard that aggregates key metrics and trends, allowing users to view overall
market conditions at a glance. Heatmaps are employed to display geographical distributions of
property prices, highlighting areas with higher or lower price ranges across cities. Bar charts
and pie charts are used to compare average property prices between cities or property types,
making it easy to identify which markets are more expensive or affordable. Line graphs track
price trends over time, providing insights into how prices have evolved and allowing users to
spot emerging trends. Scatter plots reveal correlations between property features, such as size
or age, and prices, helping to identify which attributes significantly impact value.

5.2 Screenshots

Fig 5.2.1 Home Page

Description: The home page of Real Estate Co. features a sleek property search interface
where users can input their desired city, price range, and area specifications to find suitable
housing options. The background showcases a blurred image of modern high-rise buildings,
emphasizing a focus on urban real estate.

Dept. of CS&E, MIT Mysore. 12

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Fig 5.2.2 Property search Page

Description: The property search page allows users to input specific criteria such as city,
minimum and maximum price, and minimum and maximum area in square feet. This
streamlined interface ensures users can efficiently find properties that match their
requirements, with a simple and clean design for ease of use.

Fig 5.2.3 Property Details

Description: The property details page displays a comprehensive list of available properties
matching the search criteria, including key information such as city, status, price, area, number
of bedrooms, and type of property. Each entry also provides a link for more detailed
information, making it easy for users to explore their options further.

Dept. of CS&E, MIT Mysore. 13

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

5.2.4 Property

Description: Varsha Balaji Park in Mumbai offers 3 BHK furnished apartments with a size of
1835 sqft, currently under construction. Priced at ₹22,098,905 (₹12,043/sqft), more details can
be found at [Project URL]

5.2.5 Property Page

Description: Balaji Park in Kharghar, Mumbai offers 2, 3, and 4 BHK apartments ranging
from 774 to 2,490 sqft, priced between ₹85 lakh and ₹2.8 crore. Under construction, these
apartments are available at ₹10,045/sqft.

Dept. of CS&E, MIT Mysore. 14

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

Chapter 6
CONCLUSION AND FUTURE WORK
6.1 Conclusion
In conclusion, the " Price Prediction for Indian Cities Housing Property using KNN
Algorithm " project delivers a comprehensive analysis of the real estate market across various
Indian cities, offering valuable insights into property pricing dynamics. By integrating
extensive data collection methods, including web scraping and API utilization, the project
provides a robust dataset that captures current and historical price trends. Advanced data
processing techniques and statistical analysis reveal key factors influencing property values,
while sophisticated predictive models forecast future trends, aiding in investment decision-
making.
The interactive visualization component, featuring dashboards, heatmaps, bar charts, and
scatter plots, allows users to explore data intuitively and make comparative assessments across
cities and property types. Enhanced by real-time data updates and secure user interfaces, the
project ensures timely and reliable information. Future enhancements, such as incorporating
additional data sources and developing mobile applications, promise to further refine the
analysis and expand accessibility. Overall, the project equips stakeholders with a detailed,
actionable understanding of the housing market, facilitating strategic planning and investment
in the diverse real estate landscape of Indian cities.

6.2 Future Enhancement

 Advanced Predictive Analytics: Incorporate sophisticated machine learning models

for improved price forecasting and trend analysis.
 Integration of Additional Data Sources: Include data from social media, economic
indicators, and infrastructure developments for a more comprehensive market view.
 Enhanced User Personalization: Develop features for personalized notifications and
recommendations based on user preferences and behavior.
 Geospatial Analysis: Utilize advanced mapping and geospatial analysis to visualize
property prices in relation to amenities and infrastructure.
 Mobile Application Development: Create a mobile app for accessing property price
data, interactive maps, and real-time updates on the go.
 Real-Time Data Feeds: Implement real-time data integration to ensure up-to-date
pricing information and market insights.

Dept. of CS&E, MIT Mysore. 15

 Price Prediction for Indian Cities Housing Property using KNN Algorithm

REFERENCES

[1] V. Singh, A. Sharma, and R. Kumar, "Housing Price Prediction in Indian Cities Using
Machine Learning Algorithms," 2024 IEEE International Conference on Big Data Analytics
and Computational Intelligence (ICBDACI), pp. 152-157, 2024.
[2] S. Patel, K. Joshi, and M. Dave, "Predictive Analytics for Real Estate Price Estimation in
Major Indian Cities," Journal of Real Estate Research, vol. 45, no. 3, pp. 345-360, 2023.
[3] H. Verma, R. Mehra, and P. Singh, "Developing a Robust Model for Housing Price
Prediction Using Multiple Regression Techniques," International Journal of Advanced
Computer Science and Applications (IJACSA), vol. 14, no. 2, pp. 233-240, 2023.
[4] A. Gupta, N. Sharma, and M. Rao, "Utilizing Kaggle Datasets for Accurate Housing Price
Forecasting in Urban India," 2023 IEEE International Conference on Data Science and
Advanced Analytics (DSAA), pp. 122-129, 2023.
[5] R. Roy, P. Agarwal, and S. Tiwari, "Comparative Analysis of Machine Learning Models
for Housing Price Prediction," 2024 International Conference on Artificial Intelligence and
Machine Learning (ICAIML), pp. 101-108, 2024.
[6] M. Nair, K. Bhatia, and S. Srivastava, "Incorporating Location-Based Features for
Enhanced Property Price Prediction," 2023 ACM Conference on Data Science and Machine
Learning (DSML), pp. 67-74, 2023.
[7] N. Jain, S. Sinha, and A. Saxena, "Price Trends and Forecasting in the Indian Real Estate
Market Using Deep Learning Techniques," Journal of Machine Learning and Applications, vol.
12, no. 4, pp. 567-578, 2023.

Dept. of CS&E, MIT Mysore. 16