Research Projects
Semantic Annotation
Impressions Video
Large Scale Logical Retrieval
Overview
This project is about applying logical retrieval to patent retrieval.
Probabilistic logical retrieval is useful for complex search tasks where different search criteria, dimensions, and data sources need to be combined. After an initial search for patents about semiconductors (the initial search is the basic, so-called topical search), for example, the task is to explore the patents of the inventors who filed the most relevant patents about semiconductors. Logical retrieval supports the reasoning process, which is the post-processing of retrieval results and the combination of evidence. At large, this supports the decision-making processes that patent searchers are faced with.
Large-Scale Logical Retrieval (LSLR) enables searching for patents and related entities in both full-text and Structured Query Language (SQL) fashion. Particular to LSLR is the usage and modelling of techniques from distributed information retrieval in order to scale logical retrieval to millions of patents.
Goals
The goal of this project is to investigate both the applicability and benefit of logical retrieval for patent search. In particular, the project aims to investigate the scalability of logical retrieval, where scalability is investigated with respect both to data volume and "semantic" expressiveness. Eventually, a large scale logical retrieval (LSLR) framework on a super computing infrastructure will have been developed.
Expected outcome for IP experts
Seamless searching over patents' full text and the aggregated information sources associated with the patent document.
Timeline
This project started at the end of 2008. A large scale logical retrieval framework is expected by the end of 2009, an adapted version for semantic retrieval by April 2010.
Project Partners
- University of Glasgow, UK (Research & Development)
- Queen Mary University London, UK (Research & Development)
- Matrixware Information Services GmbH, AT (showcase, funding)
- Information Retrieval Facility, AT (data, infrastructure)
Links
Matrixware.net/Logical Retrieval (for more information about methods and findings, as well as publications and related works)
Contact
Besides patent search, logical retrieval processes can be applied to various aspects of information management/retrieval. The IRF can provide you with more details about how logical retrieval processes can help in addressing your concrete needs. Please send your inquiry to: science@ir-facility.org.
