Extraction of Information from Scientific Documents for Food and Safety Assessment

In today’s rapidly evolving world, ensuring the safety of food and feed has become an indispensable priority. This report explores the utilization of ABBYY software tools and fine-tuned transformer-based language models to streamline the process of extracting critical information from scientific documents in the domain of food and feed safety assessment. The objective is to improve the effectiveness of safety assessment by integrating cutting-edge analytical methods and advanced natural language processing techniques. The study showcases the exceptional performance of the Generative Pre-trained Transformer-2 (GPT-2) model in comprehending complex scientific texts and extracting pertinent information related to analytical methods. By leveraging this model, Innovamol Consulting Srl, a specialized scientific research and development company, significantly accelerated the extraction process, saving substantial time. This research was conducted as part of an internship at Innovamol Consulting Srl, where the focus lies on chemical experimental design.