A study on learning-augmented k-means clustering

Clustering is a practical approach for extracting meaningful information from unstructured data. With the exponential growth of data, it is essential to develop efficient methods for computing clusters. Utilizing existing knowledge through prediction algorithms can significantly improve clustering techniques. By incorporating predictive models into the clustering process, we can leverage known information to enhance cluster analysis and achieve more accurate and insightful results. This study investigates the potential of learning-augmented clustering, specifically focusing on the widely used k-means algorithm. Through empirical analysis, we examine the impact of integrating prediction algorithms on clustering performance and quality.