You are on page 1of 9

House Price Prediction

using machine learning

By -
Priyanshu Shawarn – 22BET10134
Nitin Kumar Rai – 22BET10127
Dhruv – 22BET10129
Abhishek Bhadauriya – 22BET10131
Aditya Soni – 22BET10130
Introduction to House Price
Prediction using machine
learning
The need to model house price forecast has arisen from the annual
increase in property prices.
This project offers an overview of the many regression techniques that
may be used to anticipate house costs with the help of Python tools.
The more accurate prediction was produced by the suggested method,
which took into account the more sophisticated factors employed in
the property price calculation. It also gives a quick overview of the
several numerical and graphical methods needed to forecast a home's
price. This project explains what the machine learning-enabled house
price model is, how it operates, and which datasets are included in our
suggested model.
Timeline of the Reported Problem
1 Discovery
The problem was first identified and reported to the relevant stakeholders.

2 Investigation
Detailed analysis and research were conducted to understand the root
cause.

3 Resolution
Efforts were made to address the issue and implement a sustainable
solution.
Existing Solutions
Zillow's Zestimate Redfin's Home Value Tool Trulia's Trulia
Estimates
1 Zestimate takes into
2 Redfin, another real 3
Trulia, a real estate
account various estate platform, utilizes
website, employs
features such as machine learning
machine learning to
location, size, and algorithms to estimate
generate property value
historical sales data. home values. Their tool
estimates. The Trulia
The model continually considers factors such
Estimates tool takes
learns and updates its as property details,
into account property
predictions as new data location, and market
details, local market
becomes available. trends to provide users
trends, and other
with accurate and up-
relevant features to
to-date price estimates.
predict house prices.
Bibliometric Analysis
Define Research Keywords: Select Databases: Search and Retrieve Publications:

Identify keywords related to Utilize academic databases such Conduct a systematic search
your topic, such as "house price as PubMed, IEEE Xplore, using your keywords. Retrieve
prediction," "real estate ScienceDirect, and Google relevant publications, including
valuation," and "machine Scholar. These platforms provide research articles, conference
learning." These keywords will access to a vast collection of papers, and reviews. Make sure
be used to search for relevant academic papers, conference to note the publication year,
literature. proceedings, and journals. authors, and journal/conference
information.
Review Summary

Key Findings Geospatial Analysis Temporal Dynamics:


Various studies emphasize Incorporating geospatial Recognizing temporal
the importance of data, including distance to dynamics is essential for
selecting relevant features schools, public transport, accurate predictions.
for predicting house and other amenities, Time-series models and
prices. Features such as enhances the predictive recurrent neural networks
location, square footage, power of models. (RNNs) are employed to
number of bedrooms and Geospatial analysis is capture trends and cyclic
bathrooms, and proximity crucial for capturing patterns in housing
to amenities consistently location-based variations markets over time.
emerge as influential in house prices.
factors.
Problems with Existing Solutions
Overfitting:
Overfitting occurs when a model performs well on the training data but
fails to generalize to new, unseen data. This can lead to inaccurate
predictions in real-world scenarios, especially if the model is too complex
or the training dataset is not representative of the entire population.
Data Quality Issues:
Many datasets used for training machine learning models in real estate
may suffer from data quality issues, such as missing values, outliers, or
inaccuracies. Poor data quality can impact the robustness and reliability of
the models.
Limited Generalization:

Models trained on specific geographic locations or time periods may


struggle to generalize to different regions or timeframes. Real estate
markets exhibit substantial variations, and models need to be adaptable to
diverse conditions.
Goals And Objectives

Feature Importance Understanding: Accurate Price Prediction: Improved Valuation Precision:


The main objective is to Enhance the precision and
Identify and understand reliability of property
the importance of different build machine learning
models that can accurately valuation compared to
features in influencing traditional methods. Machine
house prices. This predict the prices of
houses. The model should learning models can
understanding can provide incorporate a wide range of
valuable insights for real capture complex
relationships between features, including non-
estate professionals, traditional ones like images
helping them focus on the various features and the
target variable (house and text, leading to more
most influential factors in comprehensive and precise
property valuation. prices).
valuations.
Thank You

You might also like