Python Horse Race Outcome Predictor
Budget: $30 – $250 USD
Hello everyone!
I want to build a horse race prediction system using machine learning in Python.
The goal is to use historical race data from 1990 to 2024 to predict race outcomes and understand which factors influence performance. This project includes data cleaning exploratory analysis visualization model training and evaluation.
The dataset contains yearly files for races and horses with detailed information about performance results and race conditions.
The main goal is to predict outcomes such as win or place and the secondary goals include finding the most important features handling imbalanced classes and building a reliable model using long term historical data.
The process will include cleaning missing values normalizing fields converting categories to numbers engineering new performance based features and merging horse and race data into one dataset.
The EDA will include summary statistics correlation analysis and visualizations like histograms scatter plots box plots and heatmaps. For modeling I will test different machine learning methods such as regression random forest gradient boosting and neural networks and use cross validation to measure performance.
I will also apply techniques like SMOTE under sampling or class weight adjustments for imbalance and use feature selection and hyperparameter tuning to improve results.
This post gives a simple overview of the project plan and steps. If you are comfortable with this project and can deliver quickly feel free to get started. Thanks!
I want to build a horse race prediction system using machine learning in Python.
The goal is to use historical race data from 1990 to 2024 to predict race outcomes and understand which factors influence performance. This project includes data cleaning exploratory analysis visualization model training and evaluation.
The dataset contains yearly files for races and horses with detailed information about performance results and race conditions.
The main goal is to predict outcomes such as win or place and the secondary goals include finding the most important features handling imbalanced classes and building a reliable model using long term historical data.
The process will include cleaning missing values normalizing fields converting categories to numbers engineering new performance based features and merging horse and race data into one dataset.
The EDA will include summary statistics correlation analysis and visualizations like histograms scatter plots box plots and heatmaps. For modeling I will test different machine learning methods such as regression random forest gradient boosting and neural networks and use cross validation to measure performance.
I will also apply techniques like SMOTE under sampling or class weight adjustments for imbalance and use feature selection and hyperparameter tuning to improve results.
This post gives a simple overview of the project plan and steps. If you are comfortable with this project and can deliver quickly feel free to get started. Thanks!