Scribd
Upload
363 views

Presentation On IPL Match Winner Prediction With ML

This document describes a machine learning project that predicts the winner of an IPL cricket match based on features like the playing teams, toss winner, and venue. It discusses preprocessing the dataset, encoding categorical features, training logistic regression, SVM, decision tree and random forest models on the data, and selecting random forest as it achieved the highest accuracy of 55% based on cross validation.

Uploaded by

Debajit Deka 7019
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
363 views

Presentation On IPL Match Winner Prediction With ML

This document describes a machine learning project that predicts the winner of an IPL cricket match based on features like the playing teams, toss winner, and venue. It discusses preprocessing the dataset, encoding categorical features, training logistic regression, SVM, decision tree and random forest models on the data, and selecting random forest as it achieved the highest accuracy of 55% based on cross validation.

Uploaded by

Debajit Deka 7019
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 27

PRESENTATION ON

IPL MATCH WINNER


PREDICTION WITH
ML
Submitted by:
Deepshekhar Dey
Chandan Kumar Gupta
Himadri Sikhar Gogoi
Ritusman Kashyap Bhuyan
PROJECT OVERVIEW
 Our project is basically a prediction project
which predicts which team will win in an IPL
match provided the match details like
playing teams, toss winning team, toss
decision, and the playing venue.
 This is done with the analysis of different ML
prediction algorithms and choosing the most
efficient one for the prediction.
DATASET INFO
 Our algorithm does the prediction based on
the past data of the teams that have already
played.
 This past data is provided to the algorithm in
the form of a dataset.
 In the dataset we can see that all the
information regarding the matches are
provided.
 Our model uses these past data to predict
the upcoming match events.
 Now, in our dataset there are many
unnecessary columns, so we drop those
columns and extract only those columns
which are relevant to our prediction process.
FEATURES AND TARGET
VARIABLES IN OUR DATASET

 Here in our dataset, Team 1, Team 2, Toss


Winner, Toss Decision, and Venue are our
features.
 Winning Team is our target variable.
 First, we import all our necessary python and
ML libraries that we will use in our project.
 Pandas- Helps us with data analysis.
 Numpy- Helps us with numerical
computation.
 Seaborn and Matplotlib- Helps us with data
visualization.
 The first thing we do is check whether there
are null values in our dataset or not.
 Since there is null value in our Winner
column so, we replace those values with
“DRAW”.
 After removing all the null values from our
dataset, we extract our necessary columns
which are going to be our features and target
variables.
 We can see that there are no null values in
our dataset.
 So, we proceed further.
 Since, our dataset contains categorical data,
so we need to encode them into numerical
before feeding them into ML algorithm.
 So, we use LabelEncoder, which will encode
our categorical values into numerical values.
Encoding our teams.
 Encoding our venues.
 Encoding our “Toss Decision” column.
 After encoding all categorical columns, our
dataset looks like this.
 So, we are ready to feed this dataset into ML
algorithm.
 Before feeding our dataset into our ML
model, we divide it into train data and test
data.
 Our train feature data.
 Our test feature data.
 Our train target data.
 Similarly we have our test target data.
LOGISTIC REGRESSION

 Using Logisitic Regression model in our


dataset with accuracy 30%.
SUPPORT VECTOR MACHINE

 Using Support Vector Machine(SVM) model in


our dataset with accuracy 38%.
DECISION TREE

 Using Decision Tree model in our dataset with


accuracy 48%.
RANDOM FOREST

 Using Random Forest model in our dataset


with accuracy 51%.
CROSS VALIDATION

Cross-validation is a re sampling method that uses different


portions of the data to test and train a model on different
iterations.
 After applying cross validation scores on
various algorithms on our dataset, we can
see that the Random Forest Algorithm is
providing us the maximum accuracy upto
55%.
 So, we will use Random Forest model to
predict our target variable.
 Our prediction function.
PREDICTIONS
THANK YOU!!!!

You might also like