Interpretable unsupervised decision trees
US7542960B2 · kind B2 · utility
10Cited by
11References
9Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Dec 17, 2002 |
| Grant date | Jun 2, 2009 |
| Priority date | — |
| Expiry date | Dec 31, 2024 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F18/24323
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
An unsupervised decision tree is constructed, involving the data records or patterns that do not posses any class labels. The objective of clustering or segmenting the data set is to discover subsets of data records that possess homogeneous characteristics. In the context of clustering, namely grouping or segmenting data sets without any supervised information, an interpretable decision tree is recognized as beneficial in various contexts such as customer profiling, text mining, and image and video categorization.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.