Scribd
Upload
215 views24 pages

Ai PPT

This document presents a project on developing a sports win predictor using artificial intelligence. It aims to build a model to predict the probability of winning or losing a match based on features extracted from previous match results. The project uses logistic regression for classification to calculate the weightage of each feature and predict the target variable. It extracts features from training data, develops the model, obtains current match features and predicts the win percentage. The applications include predicting winners in sports leagues. Future work is to enhance the model accuracy by increasing the dataset size.

Uploaded by

Abhishek Koushik
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
215 views24 pages

Ai PPT

This document presents a project on developing a sports win predictor using artificial intelligence. It aims to build a model to predict the probability of winning or losing a match based on features extracted from previous match results. The project uses logistic regression for classification to calculate the weightage of each feature and predict the target variable. It extracts features from training data, develops the model, obtains current match features and predicts the win percentage. The applications include predicting winners in sports leagues. Future work is to enhance the model accuracy by increasing the dataset size.

Uploaded by

Abhishek Koushik
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

Department of Computer Science and Engineering

BANGALORE INSTITUTE OF TECHNOLOGY


Artificial Intelligence, Fifth Semester 2019-20 (odd)

SPORTS WIN PREDICTOR

Presented by

Abhishek Koushik B N: 1BI17CS005


Akash R : 1BI17CS010
Anandteerth : 1BI17CS015
Prajwal V G : 1YD17CS011

Under the guidance of,


Dr. Suneetha K R
Associate Professor,
Dept. of CSE, BIT
AGENDA
 Introduction
 Problem Statement
 Objectives
 Architecture
 Technology / Algorithm's used
 Implementation Details
 Results
 Applications
 Conclusion and Future work
 References
INTRODUCTION
 We have all seen sports matches where the outcome of the match is
predicted beforehand. This is done using Artificial Intelligence. In
this project, we target to build our own model for predicting wins in
a particular tournament.
 The data used for developing the model is factors on which win
depends from previous tournaments. This is known as Classification
problem in AI. We use methods like Logistic Regression to predict
the probability percentage of win or lose.
 We try to fit a linear curve to the given data set with the
optimal training error by associating optimal weights to the features.
Then we take the present value and try to find the target value i.e.
the percentage probability of Win or lose by giving the values of
features to the equation.
PROBLEM STATEMENT

 “To predict Win/Loss percentage for a given match in a sport using


the results of previous matches using Artificial Intelligence.”
 Input: Dataset with features such as Run Rate, Net Run
Rate, Number of goals, Goal Difference, etc which
decide the outcome of the match.
 Output: Confidence of Win/Loss based on present
features.
OBJECTIVES
 Collection of results of previous matches
 Extraction of features that determine a win for a team
 Calculate the weightage of each feature to Target i.e. Win percentage
 Develop the model which fits the data
 Obtain present set of features
 Predict the win percentage using the model developed before and
present set of features.
ARCHITECTURE
Y^ = W0 + W1(#Feature 1) + W2(#Feature 2) + …
Where,
W0 -> Intercept
W1, W2, … -> Weights of Features
 The Data is divided into Training and testing data as a random split.
The training data is used to train the model and testing data is used
to evaluate the trained model as a proxy.
 The training data is then subjected to Feature Extraction where we
derive the set of features on which the target prediction depends on.
 The Machine Learning algorithm then predicts the best weights for
each feature and in each iteration, it becomes more accurate. We stop
the training process when the required accuracy is reached.
ARCHITECTURE

Data Set

Training Data Testing Data


TECHNOLOGY / ALGORITHM'S USED
Technologies:
 Python
 TuriCreate
 Google Co-Laboratory
 Anaconda
Algorithm:
 Logistic Regression algorithm for Classification
IMPLEMENTATION DETAILS
Training

X Y^
Data

Feature Logistic
Extraction Classifier

W^
Y

Optimizer

Quality
Metric
Win

Current
Logistic Classifier
Features

Lose
RESULTS
APPLICATIONS
 Sports league winner prediction
CONCLUSION AND FUTURE
ENHANCEMENT
 As the size of the dataset increases the accuracy of the model also
increases.
 Thus, More the accuracy of model, better will be the predictions
made.
REFERENCES
1. Machine Learning Foundations: A Case Study Approach by
University of Washington.
THANK YOU

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy