一种高维数据聚类方法、装置、设备、介质及产品
By constructing local information factors to update the membership and similarity matrices, and combining this with projection matrix optimization, the problem of inaccurate clustering results in the single-stage method is solved. This achieves the preservation of the local neighborhood structure of high-dimensional data in a low-dimensional space, thereby improving the clustering accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAMEN UNIV
- Filing Date
- 2024-08-26
- Publication Date
- 2026-07-17
AI Technical Summary
Single-stage methods have low clustering accuracy in high-dimensional data clustering, and existing methods ignore the local neighborhood structure of high-dimensional data, resulting in inaccurate clustering results.
By constructing a local information factor based on high-dimensional data samples to update the membership matrix, combining the projection matrix and cluster pair similarity matrix, constructing an auxiliary matrix and calculating eigenvalue decomposition, updating the projection matrix, constructing a convex optimization problem, updating the cluster pair similarity matrix, preserving the local neighborhood structure of high-dimensional data samples, and combining dimensionality reduction and clustering optimization.
It preserves the local clustering structure of high-dimensional data in a low-dimensional space, improves the accuracy of clustering results, and effectively clusters high-dimensional data.
Smart Images
Figure CN119128566B_ABST