New Method for Keyword Extraction for Patent Claims

TL;DR

This paper introduces a novel keyword extraction method tailored for patent claims, aiming to improve prior art search accuracy by leveraging the unique information structure of patent claims.

Contribution

The paper presents a new keyword extraction technique specifically designed for patent claims, differing from traditional frequency-based methods.

Findings

The new method outperforms frequency-based approaches in relevant patent retrieval.

It effectively captures important keywords from patent claims.

The approach enhances the accuracy of prior art searches.

Abstract

The search for prior art is crucial in patent application processing, it consists in retrieving other documents relevant to the invention of the application. Most methods feed a search engine with keywords that are extracted by frequency-analysis methods. We suggest and demonstrate a new method that relies on the way information is provided in patent claims.