Proceedings of the 2nd international workshop on Patent information retrieval

Abstract

We welcome you to PaIR 09, the 2nd workshop on patent information retrieval, organised by the Information Retrieval Facility (IRF) and Matrixware Information Services. This second edition of our workshop extends and continues topics covered in 2008, but also addresses new interesting subjects. Patent Information Retrieval is an active and challenging field both for researchers and for professional information specialists. Patents play a key role not only in protecting intellectual property but also as a strategic business factor in all modern economies. Despite the enormous advances in Information Retrieval techniques in the past few years, advanced search tools for patent professionals are still in their infancy, so the research in patent retrieval represents an important opportunity for research. Patent search is a particular challenge to information retrieval and access systems. Amongst the challenges successful patent search of search system of the future will have to face are; very large numbers of highly complex structured documents;highly heterogeneous document collections (scientific papers, legal public disclosure as well as patents);multiple languages;ambiguous and conflicting jargon; complex languagetechnological concepts;sophisticated legal jargon;ranges and other complex query formstracking temporal issues like publication datatabular and graphical information mixed into text;and so on. The objective of the workshop is to provide a forum for Information Retrieval and Knowledge Management scientists as well as Patent Retrieval experts from industry to study the next generation of patent search tools. This year the workshop received 13 submissions, of which it accepted 5 full papers. An additional 6 papers were invited to do a short presentation and display posters, in order to trigger interesting discussions on future directions of Patent Information Retrieval. The 5 full papers cover some of the most salient aspects of Patent IR. Phrase-based Document Categorization Revisited (Koster and Beney) addresses the problem of whether or not the use of linguistic techniques improves patent classification. In Identification of Low/High Retrievable Patents using Content-Based Features, Bashir and Rauber analyse the factors that lead to bias in retrieval systems in order to attempt to compensate and make all documents 'retrievable'. Yang and colleagues (A Design Rationale Representation Model using Patent Documents) show us how to model and extract specific information (i.e. design rationales) from patent documents. In a similar line of thought, Tiwana and Horowitz (Extracting Problem Solved Concepts from Patent Documents) work on extracting the essence of a patent: the problem being solved by that specific invention, in order to improve future prior art searches. Further improvements in patent searches may be obtained if one follows the ideas suggested by Klampanos and colleagues in A Case for Probabilistic Logic for Scalable Patent Retrieval. In addition to the 5 full papers, 6 research groups will describe their ideas in a booster session, followed by poster presentations during the workshop's break. The topics covered by these papers tackle patent classification and categorisation, translation, natural language processing and, of course, new ways of performing prior art search. We expect that these papers will trigger interesting conversations and future development in IP search, through a closer collaboration between researchers and industry representatives. We also hope that the workshop will be a springboard for many future events and lead to the recognition of patent searching as one of the central areas of research in Information Retrieval and Knowledge Management.