Density Peak Clustering on Large and Dynamic Data

Clustering groups together objects into clusters such that the intra-cluster similarity is higher than the inter-cluster similarity and helps in data analysis. However, for large data, the clustering techniques may take several hours and days to report the clusters. Moreover, many real-world applications deal with data that is frequently updated. For e.g., users posting tweets on social media. Conventional algorithms designed for static data are not suitable for such tasks. Therefore, this thesis design and develop solutions for density-peak clustering technique to (i) quickly find clusters for large data, and (ii) quickly maintain the clusters when the data is updated.