SCIENTIFIC OBJECTIVES

PatExpert’s overall scientific objective is to change the paradigm currently followed for patent processing from textual (viewing patents as text blocks enriched by picture material that are interpreted as “canned” material, sequences of morpho-syntactic tokens, or collections of syntactic structures) to semantic (viewing patents as multimedia knowledge objects) processing. The overall goal translates into nine more specific scientific objectives:

[1] [2] [3] [4] [5] [6] [7] [8] [9]

[9] To develop patent valuing and technology area assessment techniques that will offer an adequate alternative to manual techniques used so far for this purpose.
We will target machine learning-based techniques for estimating the technical value of patent applications and registered patents, patent collating, the patent generating potential of specific technology areas by semantic clustering – either in general or with respect to a particular topic –, etc. In relation to these techniques, we will explore, among other parameters, abstraction estimation, use of patent holder/claim profile data (such as number of patents, accounting information, etc.), use of filing attorney profile information, use of information on the number and type of scientific publications in the area or on similar topics.
Addressing this objective, PatExpert places itself at the head of the research on patent processing.

TECHNOLOGICAL OBJECTIVES

PatExpert’s overall innovation-related technological objective is the development of a functioning service. The service is targeted as a showcase that demonstrates the viability of all techniques developed in PatExpert for real applications:

[1] [2] [3] [4] [5]