Document summarization based on topicality and specificity
US7346494B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Oct 31, 2003 |
| Grant date | Mar 18, 2008 |
| Priority date | — |
| Expiry date | Apr 10, 2026 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F40/289
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Topicality scores are determined for a number of phrasal expressions in documents. Phrasal expressions may be noun phrases, with or without corresponding prepositional phrases, subject-verb pairs, and verb-object pairs. The documents describe some topic or multiple topics. Techniques can be used to determined how the phrasal expression compares with the topic or topics being described in the documents. Specificities are determined for the phrasal expressions. Techniques may be used to determine whether phrasal expressions are more or less specific than other phrasal expressions. An order is determined for the phrasal expressions by using the topicality scores and the specificities. The order may be represented as a phrasal expression tree, for example. The phrasal expression tree may be displayed to a user, and the user can navigate through the phrasal expression tree, and therefore through the one or more documents.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.