Natural Language Processing (NLP) – A Solution for Knowledge Extraction from Patent Unstructured Data

Abstract

Patents are valuable source of knowledge and are extremely important for assisting engineers and decisions makers through the inventive process. This paper describes a new approach of automatic extraction of IDM (Inventive Design Method) related knowledge from patent documents. IDM derives from TRIZ, the theory of Inventive problem solving, which is largely based on patent's observation to theorize the act of inventing. Our method mainly consists in using natural language techniques (NLP) to match and extract knowledge relevant to IDM Ontology. The purpose of this paper is to investigate on the contribution of NLP techniques to effective knowledge extraction from patent documents. We propose in this paper to firstly report on progress made so far in data mining before describing our approach.