MGNR: A Multi-Granularity Neighbor Relationship and Its Application in KNN Classification and Clustering Methods

Jiang Xie,Xuexin Xiang,Shuyin Xia,Lian Jiang,Guoyin Wang,Xinbo Gao
DOI: https://doi.org/10.1109/tpami.2024.3400281
IF: 23.6
2024-01-01
IEEE Transactions on Pattern Analysis and Machine Intelligence
Abstract:In the real world, data distributions often exhibit multiple granularities. However, the majority of existing neighbor-based machine-learning methods rely on manually setting a single-granularity for neighbor relationships. These methods typically handle each data point using a single-granularity approach, which severely affects their accuracy and efficiency. This paper adopts a dual-pronged approach: it constructs a multi-granularity representation of the data using the granular-ball computing model, thereby boosting the algorithm's time efficiency. It leverages the multi-granularity representation of the data to create tailored, multi-granularity neighborhood relationships for different task scenarios, resulting in improved algorithmic accuracy. The experimental results convincingly demonstrate that the proposed multi-granularity neighbor relationship effectively enhances KNN classification and clustering methods. The source code has been publicly released and is now accessible on GitHub at https://github.com/xjnine/MGNR .
What problem does this paper attempt to address?