A Machine learning Classifier coded in python that uses k-means clustering and k-nearest neighbor machine learning algorithms to classify whether or not a given NBA player's statisics should be an all star.
NBA-AllStar-Classifier
The NBA-AllStar-Classifier is a machine learning project designed to classify NBA players as All-Stars or non-All-Stars based on their performance metrics. The classifier utilizes two popular machine learning techniques: k-means clustering and k-nearest neighbor (k-NN) algorithms.
Data Clustering with k-means The system begins by applying k-means clustering to group players into clusters based on key performance statistics such as points, rebounds, assists, steals, blocks, and more. This unsupervised step helps identify natural groupings of players with similar playing styles and statistical profiles.
Classification with k-nearest neighbor (k-NN) After clustering, the classifier employs the k-nearest neighbor algorithm to predict All-Star status. The k-NN model is trained on historical data of NBA players labeled as All-Stars or non-All-Stars. Given a new player's performance data, the classifier finds the k most similar players and predicts whether the player should be an All-Star based on the majority class of these neighbors.
The combination of k-means clustering for initial grouping and k-NN for final classification ensures that players are evaluated not just in isolation, but also in the context of similar players. This hybrid approach leverages both unsupervised and supervised learning to provide a robust and data-driven method for identifying NBA All-Star talent.