A Fast Partitional Clustering Algorithm based on Nearest Neighbours Heuristics

•Investigate K-means clustering for large collections of sparse vectors of high dimensionality.•Proposed utilizing the inverted list data-structure to improve run-time of K-means.•Heuristics proposed for initial centroid selection and centroid updates.•Proposed approach outperforms the run-time of K...

Ausführliche Beschreibung