CricStrategizer 🚀🏏

Overview

Cricket, the second most popular sport globally with 2.5 billion fans, is witnessing continuous growth in popularity. The Indian Premier League (IPL), valued at 5.3 billion in 2017, showcases the sport's immense appeal. Viacom18 secured IPL digital rights for 3.0 billion, while Disney Star retained TV rights for 3.05 billion for 2023-27. With 60+ national teams and professional players, cricket is a dynamic and exciting sport with a worldwide fan base actively playing and investing their time.

Project Objective

CricStrategizer is designed to revolutionize cricket strategy by predicting a batsman's shot outcome for a particular delivery based on specific properties. By leveraging machine learning models developed on IPL 2022 data, the project aims to empower players and teams to strategically analyze and understand a batsman's playing style, facilitating effective game plans and strategies.

Project Workflow

1. Data Extraction and Cleaning: Raw data is processed for cleanliness and readiness.
2. EDA and Visualizations: Exploratory Data Analysis and visualization for pattern identification.
3. Feature Engineering: Enhances predictive capabilities by engineering important features.
4. Model Creation: Develops SVM, Random Forest, and ANN models optimized for predicting shot outcomes.

How to Use

1. Clone the repository:

   git clone https://github.com/syam-m/CricStrategizer.git
   
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Execute the DAG file:

   python dag.py
   

Future Enhancements

1. Implement real-time data integration for up-to-the-minute analysis.

   • Enhance the system to dynamically integrate real-time data, providing the latest insights for strategic decision-making.

2. Explore additional machine learning models for further accuracy improvement.

   • Investigate and incorporate advanced machine learning models to continually enhance the accuracy and predictive capabilities of CricStrategizer.

3. Develop a user interface for easy interaction with predictive models.

   • Create a user-friendly interface that allows cricket enthusiasts, players, and teams to easily interact with and benefit from the predictive models, making the tool accessible and intuitive.