Information processing device, information processing method, and program
By setting two thresholds in the hierarchical clustering technology and selectively adding data points with specific similarity, the problem of noise pollution in clustering is solved and the purity and accuracy of clustering is improved.
JP7654421B2Active Publication Date: 2025-04-01CANON KK
4 Cites 0 Cited by
Patent Information
- Application Number
- JP2021028715
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-02-25
- Publication Date
- 2025-04-01
- Estimated Expiration
- 2041-02-25
AI Technical Summary
Technical Problem
In the prior art, there is a problem of noise pollution in the cluster generated by hierarchical clustering technology, especially in clusters with large amounts of data, it is difficult to effectively reduce noise.
Method used
By setting two different thresholds (TH1 and TH2), during the cluster generation process, data points with high similarity are clustered and data points with specific similarity are added to the cluster to reduce noise.
Benefits of technology
It effectively reduces the existence of noise in clustering, improves the purity and data volume of clustering, and increases the accuracy of clustering results.
✦ Generated by Eureka AI based on patent content.
Abstract
To reduce noise contained in a cluster in which plural pieces of data are classified.SOLUTION: An information processing device includes: cluster obtaining means for obtaining at least one cluster from plural pieces of data on the basis of a first threshold; and deciding means for deciding data to be added to the cluster on the basis of a similarity between the plural pieces of data and data contained in the cluster, and of a second threshold different from the first threshold.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art