Research Projects
Semantic Annotation
Impressions Video
Searching for Prior Art Patents
Overview
The task of searching for related patents (or literature) is far more complicated than performing a simple web search. The patent is analysed in order to build a feature set that is used to search for related patents and open source literature in huge information repositories.
Goals
The goal of this project is to study automated methods for finding related patents based on a combination of many features extracted from the target patent and, thus, allow to address the task of finding prior art for a patent (application). In this setting, the patent document itself is used as the query for finding prior art. Retrieval is accomplished by building feature sets including languages models, n-grams, etc. of both the query patent and the patents in a large patent corpus. The approach employs ranking algorithms such as RankNet to discover highly associated resources.
Expected outcome for IP experts
A concept for a novel and powerful prior art search engine.
Timeline
The project has started in 2008. Design recommendations for a prior art search tool are expected by summer 2009.
Project Partners
- University of Massachusetts Amherst , US (Research & Development)
- Matrixware Information Services GmbH, AT (showcase, funding)
- Information Retrieval Facility, AT (data, infrastructure)
Links
Matrixware.net/Prior Art Patents (for more information about methods and findings, as well as publications and related works)
Contact
The IRF can provide you with more details about how this project can help in addressing your concrete needs. Please send your inquiry to: science@ir-facility.org.
