Application of state-of-the-art NLP models increases efficiency of third party patent monitoring in the nutrition and bioscience industry.
Testimonial
Objective
Identifying relevant third-party patents using transformer-based classification models.
Background
Every year millions of patents are being published worldwide covering a vast variety of topics. Patent applications generally average ~10,000 words using unique, highly context dependent, meticulously wordsmithed language (aka “legalese” or “attornish”). Monitoring third party patents is a crucial element of business development and innovation for many companies.
Keyword-based search strategies can help to reduce screening efforts by subject matter experts (SMEs). However, even with a highly customized framework of rules it is challenging to make a selection containing mainly relevant patents. This results in a substantial time investment to manually screen irrelevant patent documents.
Findings
The Institute of Data Science FHNW successfully developed a transformer-based classification model ensemble trained on third party patents annotated by DSM SMEs. A field study revealed that this model allows more efficient patent screening reducing substantially labor costs. Moreover, the model allows the pool of patents screened for relevance to be expanded, hence enabling identification of additional potentially relevant patents. Based on the PoC success, DSM intends to implement the solution on premise as a next step.
Project details
- Research areas
- AI, Machine Learning & Natural Language Processing (NLP)
- Topics
- Data Science und Engineering and Informatik und Data Science
- University
- Hochschule für Informatik FHNW, FHNW School of Computer Science / Institute of Data Science
- Partner
- DSM Nutritional Products Ltd.
- Running time
- 6 Monate
- Management
- Prof. Dr. Daniel Perruchoud,
- Collaboration
- Dr. Fernando Benites, Dominik Frefel, Joshua Meier
Contact us
For further information about the FHNW School of Computer Science or to discuss potential collaboration opportunities, please contact us.
More Projects
DrugSafety: Semi-automated reporting of side effects of drugs
- Institute
- School of Engineering and Environment
Marvel: Real-time pollen information
- Research field
- AI, Machine Learning & Natural Language Processing (NLP), Exploratory Data Science and Image Processing & Computer Vision
Knowledge Assistant – AI-based information retrieval tool
- Research field
- AI, Machine Learning & Natural Language Processing (NLP)