International Journal of Scientific Research in Computer Science, Engineering and Information Technology

ISSN : 2456-3307 (www.ijsrcseit.com)

doi :https://doi.org/10.32628/CSEIT2390359

IPL Data Analysis and Prediction Using Machine Learning

Dr. Vilas Joshi1, Yash Padol2, Kaiwalya Sevekar2, Chandan Sinha2, Rizwan Shaikh2
1Assistant Professor, Department of Computer Engineering, ISBM College of Engineering, Pune, Maharashtra,

India
* Student, Department of Computer Engineering, ISBM College of Engineering, Pune, Maharashtra, India
2

ARTICLEINFO ABSTRACT

The Indian Premier League (IPL) is a highly attended T20 cricket league
Article History:
founded by the Board of Control for Cricket in India (BCCI) in 2008.
Accepted: 01 May 2023 Valued at $6.7 billion in 2019, the IPL remains the most popular cricketing
Published: 20 May 2023 event in the world, even amidst the COVID-19 pandemic. Cricket is a
beloved sport in India and enjoys significant media attention and financial
investment. With the rise of technology, audiences can access affordable
Publication Issue live streaming of matches from anywhere. Cricket analytics, powered by
Volume 9, Issue 3 tools like NumPy, allow for the use of data to improve performance, study
May-June-2023 business opportunities, and provide insights and predictive intelligence on
game outcomes.
Page Number
205-208

I. INTRODUCTION classification are two types of supervised learning.

Unsupervised learning involves allowing the model to
The Indian Premier League (IPL) is a T20 cricket work on its own to discover patterns in unlabelled
tournament held annually in India from April to May, data. In contrast, reinforcement learning focuses on
with 8 teams representing different states. It is the learning through trial and error to maximize rewards.
most attended cricket league globally, and in 2010, it
became the first sporting event to be broadcasted live. II. LITRATURE SURVEY
This paper examines a prediction model for IPL
outcomes, which uses SVM and Logistic Regression, G.Sudhamathy's research paper helps to understand
with runs scored by the batsman in the previous ball the working principle of various machine learning
as the labelled data. algorithms and their implementation. It creates a
Machine learning techniques like supervised, model and training dataset and helps make
unsupervised, and reinforcement learning algorithms predictions using the model created. The model
are used to predict or classify information. Supervised classifies the data and compares the results to get the
learning is the most popular type and involves accuracy that matters.
learning a function that maps input to output based As in a dataset, there are many parameters. Which
on example input-output pairs. Regression and parameters are useful in the project. Factors

Copyright: © the author(s), publisher and licensee Technoscience Academy. This is an open-access article distributed 205
under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-
commercial use, distribution, and reproduction in any medium, provided the original work is properly cited
 Dr. Vilas Joshi al Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., May-June-2023,9 (3) : 205-208

influencing the concept has been taken by formatted way every time you work with it. For this
Maheshwari in his live cricket score prediction from we use a data pre-processing task.
which we come to know the main factors which are Pre-processing is a machine learning term that refers
required for score prediction and winning team to the transformation of raw features into data that
prediction. The role of classification is elucidated in machine learning algorithms can understand and
the work of Tejinder Singh, which provides correct learn from.
information or the use of naive bias and linear
regression. They provide proper knowledge of data B. Feature Extraction:
collection and preparation as well as how to train the Feature extraction aims to reduce the number of
data and test the data they provide which is more features in a data set by generating new features from
useful. existing features and then discarding the original
The brief idea of Support Vector Machines is taken features. This reduced new feature set should
from Aminul Islam Anik's paper which is about summarize most of the information contained in the
player performance in this article, the idea of SVM original feature set. Thus, a condensed version of the
system is given in detail where the prediction of original function can be created from the combination
player performance by collecting old information or of the original set.
data is given. From the literature survey, machine
learning requires prediction. C. Classifier Training:
In data science, a classifier is a type of machine
III.SYSTEM ARCHITECTURE learning algorithm used to assign class labels to data
inputs. An example is an image recognition classifier
to label images (eg. "car", "truck" or "person").
Classification algorithms are trained using labelled
data in an image recognition example, eg a classifier
that receives training data to label images. After
adequate training, the classifier can take unlabelled
images as input and generate classification labels for
each image. Classification algorithms use sophisticated
mathematical and statistical methods to generate
predictions about the probability that a data record
Fig 1. System Architecture will be classified in a certain way. In the image
recognition example, the classifier statistically
IV. MODULES predicts whether the image is likely to be a car, truck,
or person, or some other classifier that the classifier
A. Pre-processing: was trained to identify.
Data pre-processing is the process of preparing raw
data and fitting it for machine learning models. This is D. Model Training::
the first critical step in building a machine learning A logistic regression model is trained on the pre-
model. When creating machine learning projects, you processed data to predict the winning team based on
don't always come across clean and formatted data. It the available features. The training dataset is split into
is also necessary to save the data in a clean and training and validation sets to tune the model's
hyperparameters and avoid overfitting.

Volume 9, Issue 3, May-June-2023| http://ijsrcseit.com 206

 Dr. Vilas Joshi al Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., May-June-2023,9 (3) : 205-208

Analyzing past IPL matches: The system should be

E. Testing: able to analyze past IPL matches and gather data
The process of training an ML model involves related to the teams, players, venues, and other
supplying the ML algorithm (ie, the learning relevant factors.
algorithm) with training data to learn from. The term Identifying key factors for winning: The system
ML model refers to the model artifact generated by should be able to identify key factors that contribute
the training process. The training data must contain to the success of a team in IPL matches, such as the
the correct answer, called the target or target team's past performance, player statistics, pitch
attribute. A training algorithm finds patterns in the conditions, weather, and other factors.
training data that map attributes of the input data to a Developing a prediction model: The system should be
target (the response you want to predict) and builds able to develop a predictive model using logistic
an ML model that captures those patterns. regression that can predict the winning team for
future IPL matches.
V. MOTIVATION Evaluating model performance: The system should be
able to evaluate the performance of the prediction
The motivation for the "Indian Premier League model using accuracy, precision, recall, and F1-score
Analysis and Winning Team Prediction" project using metrics.
Logistic Regression could be several factors. One of
the primary motivations could be to analyze the VII. SYSTEM REQUIRMENT
performance of different teams and players in the
Indian Premier League (IPL) cricket tournament over A. Software Requirement
the years. Another motivation could be to predict the 1. Operating system : Windows 10.
winning team based on various factors such as the 2. Coding Language : Python
players' performance, team composition, weather 3. IDE : Spyder
conditions, and other relevant factors.
Such a project could be helpful to cricket fans, sports B. Hardware Requirement
analysts, team managers, and betting enthusiasts, 4. System : Intel I5 Processor.
among others. By analyzing and predicting the 5. Hard Disk : 40 GB.
performance of teams and players, this project could 6. Monitor: 15.
provide insights into the strengths and weaknesses of 7. Ram : 8 GB
different teams and help in making better decisions
while placing bets or drafting teams for fantasy VIII. METHODOLOGY
cricket leagues. Additionally, the project could also be
useful for team managers to analyze and improve Logistic Regression:
their team's performance based on the insights gained Logistic Regression is a statistical method used for
from the analysis. classification problems.
It calculates the probability of the outcome of a binary
variable based on input features.
In this project, Logistic Regression is used to predict
VI.OBJECTIVE OF THE SYSTEM the winning team in Indian Premier League matches.
The model is trained on historical data and tested on
new data to evaluate its accuracy.

Volume 9, Issue 3, May-June-2023| http://ijsrcseit.com 207

 Dr. Vilas Joshi al Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., May-June-2023,9 (3) : 205-208

Feature engineering is an important step in Logistic of Future Generation Communication and

Regression to select the relevant features that impact Networking Vol. 13, No. 1s, (2020), pp. 17- 23
the outcome. [6]. Ruchitha M and Dr. Preeti Savant, “IPL
WINNER PREDICTION USING ML
IX. CONCLUSION ALGORITHMS”,IJEASTVol. 6, Issue 9, January
2022
The Indian Premier League Analysis and Winning [7]. RabindraLamsal and Ayesha Choudhary,
team Prediction system using logistic regression is an “Predicting Outcome of Indian Premier League
effective tool for analyzing and predicting the (IPL) Matches Using Machine Learning”,
outcomes of IPL matches. The system was able to arXiv:1809.09813 [stat.AP] (September 2018)
achieve an accuracy of around 70% in predicting the [8]. S. Abhishek, Ketaki V. Patil, P. Yuktha and S.
winning team of IPL matches. The project Meghana,“Predictive Analysis of IPL Match
demonstrated the usefulness of machine learning Winner using Machine Learning
algorithms and their applications in the field of sports. Techniques”,International Journal of Innovative
Further improvements can be made by incorporating Technology and Exploring Engineering, Vol. 9,
additional features and data sources to enhance the No. 1, pp.430-435, 2019.
accuracy of the predictions. [9]. Vistro, Daniel, Leo Gertrude David,“The cricket
winner prediction with application of machine
X. REFERENCES learning and data analytics,”International Journal
of Scientific and Technology Research, Volume
[1]. G. Sudhamathy and G. Raja Meenakshi, 8, Issue 09, 2019.
“PREDICTION ON IPL DATA USING
MACHINE LEARNING” ICTACT JOURNAL ON
SOFT COMPUTING,OCTOBER 2020. Cite This Article :
[2]. Aminulislamanik, SakifYeaser, A.G.M. Imam
Hussain, AmitabhaChakrabarty, “Player’s Dr. Vilas Joshi, Yash Padol, Kaiwalya Sevekar,
Performance Prediction in ODI Cricket Using Chandan Sinha, Rizwan Shaikh, "IPL Data
Machine Learning Algorithms,”BRAC Analysis and Prediction Using Machine Learning",
university,2019.
International Journal of Scientific Research in
Computer Science, Engineering and Information
[3]. KilariJaswanth, KenchamArun, B Deva
Technology (IJSRCSEIT), ISSN : 2456-3307,
Deekshith, KilariAshwik, Pavithra J, “Analyzing
Volume 9, Issue 3, pp.205-208, May-June-2023.
and Predicting Outcomes of IPL Cricket
Available at doi :
Data,”IRJET Volume: 09 Issue: 03, Mar 2022.
https://doi.org/10.32628/CSEIT2390359
[4]. D. Jyothsna, K. Srikanth, “Analyzing and
Journal URL : https://ijsrcseit.com/CSEIT2390359
Predicting outcome of IPL Cricket Data”,
International Journal of Innovative Research in
Science, Engineering and Technology,Vol. 8,
Issue 4, April 2019.
[5]. SiddharthSinha, PrakashTripathi,
AakashVishwakarma, Prof. AmrutaSankhe, “IPL
Win Prediction System To Improve Team
Performance using SVM.,”International Journal

Volume 9, Issue 3, May-June-2023| http://ijsrcseit.com 208