Generating implicit labels and training a tagging model using such labels
US8250015B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Apr 7, 2009 |
| Grant date | Aug 21, 2012 |
| Priority date | — |
| Expiry date | Mar 14, 2031 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F18/29
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A training module is described for training a conditional random field (CRF) tagging model. The training module trains the tagging model based on an explicitly-labeled training set and an implicitly-labeled training set. The explicitly-labeled training set includes explicit labels that are manually selected via human annotation, while the implicitly-labeled training set includes implicit labels that are generated in an unsupervised manner. In one approach, the training module can train the tagging model by treating the implicit labels as non-binding evidence that has a bearing on values of hidden state sequence variables. In another approach, the training module can treat the implicit labels as binding or hard evidence. A labeling system is also described for providing the implicit labels.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.