A Comparative Analysis of Diverse Classification Techniques in Machine Learning for Predicting Poker Hands

This paper introduces an innovative approach to predicting poker hands through the application of machine learning algorithms. In our study, we delve into the realm of predictive analytics, exploring the effectiveness of various classification techniques on a comprehensive dataset of poker hands with known outcomes. The utilized algorithms encompass K-Nearest Neighbors (KNN), Logistic Regression, Adaboost, Decision Trees, Random Forest, LightGBM classifier, XGBoost, and Naive Bayes. Leveraging the inherent capabilities of these algorithms, we conduct an in-depth analysis of features extracted from the cards and their combinations, aiming to enhance predictive accuracy. To facilitate this investigation, we utilize a CSV file containing the pertinent data required for training and testing the models. The wide array of algorithms chosen permits a detailed assessment of their effectiveness in forecasting poker hands, taking into account measures like accuracy, sensitivity, and F1 measure. Integrating KNN, Logistic Regression, Adaboost, Decision Trees, Random Forest, LightGBM_classifier, XGBoost, and Naive Bayes offers a comprehensive understanding of the strengths and weaknesses of these algorithms in predicting poker hands. This research not only advances the field of predictive modeling in poker games but also contributes valuable insights into the optimal choice of algorithms for similar classification tasks.