TNStream: Applying Tightest Neighbors to Micro-Clusters to Define Multi-Density Clusters in Streaming Data
Qifen Zeng, Haomin Bao, Yuanzhuo Hu, Zirui Zhang, Yuheng Zheng, Luosheng Wen
TL;DR
TNStream introduces a theoretically grounded data stream clustering framework built on Tightest Neighbors (TN) and Skeleton Set concepts to define robust, multi-density clusters online. The method combines micro- and macro-clustering with an adaptive radius derived from Shared Nearest Neighbors (SNN) and employs Tightest Neighbors-based clustering (kTNC) to form macro-clusters, using LSH, KD-Tree, or Ball-Tree to accelerate neighbor queries. Key contributions include the formal TN framework, the $k$-tightest neighborhood closures ($k$-MTNCIS), the Tightest Neighbors Outlier Factor (TNOF), and a fully online TNStream algorithm that handles high-dimensional, noisy, and arbitrarily shaped data, as demonstrated on diverse synthetic and real-world datasets. Experimental results show that KD-TNStream and BT-TNStream achieve superior clustering quality and robustness to outliers, particularly in multi-density scenarios, while providing competitive runtime performance thanks to efficient neighbor search structures. Overall, the work advances data stream clustering theory and provides a practical, scalable solution for complex streaming data environments.
Abstract
In data stream clustering, systematic theory of stream clustering algorithms remains relatively scarce. Recently, density-based methods have gained attention. However, existing algorithms struggle to simultaneously handle arbitrarily shaped, multi-density, high-dimensional data while maintaining strong outlier resistance. Clustering quality significantly deteriorates when data density varies complexly. This paper proposes a clustering algorithm based on the novel concept of Tightest Neighbors and introduces a data stream clustering theory based on the Skeleton Set. Based on these theories, this paper develops a new method, TNStream, a fully online algorithm. The algorithm adaptively determines the clustering radius based on local similarity, summarizing the evolution of multi-density data streams in micro-clusters. It then applies a Tightest Neighbors-based clustering algorithm to form final clusters. To improve efficiency in high-dimensional cases, Locality-Sensitive Hashing (LSH) is employed to structure micro-clusters, addressing the challenge of storing k-nearest neighbors. TNStream is evaluated on various synthetic and real-world datasets using different clustering metrics. Experimental results demonstrate its effectiveness in improving clustering quality for multi-density data and validate the proposed data stream clustering theory.