Data gathering and bibliome curation to support biomedical intelligent algorithms for the reconstruction of knowledge related to the gluten protein

• Autor(a) principal: Ferreira, Tânia Raquel Moreira
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10348/11160
• topic: Gluten; biocuration

The increase in scientific research documents continues to grow significantly, making it very difficult to access relevant information from the domain we want. To solve this problem, several text mining techniques for automatic text processing have been developed to help researchers find information of interest and infer new theories. For this automatic document processing, the prior existence of a large amount of manually annotated documents is necessary for the development of robust text mining and machine learning models. To this work, a total of 5,300 documents related to the gluten protein were retrieved from PubMed, which were manually annotated, classified and from which relationships between the various biological entities were extracted for the creation of classification and relationship extraction models. Gluten is a protein that can be found in various foods and is related to the onset of several diseases, the most common being celiac disease. Following a gluten-free diet is currently the only known treatment for diseases related to this protein. However, nowadays more and more people follow a gluten-free diet on their initiative and without any related health issues. People are seeking answers through the internet and social networks, as there is an abundance of information, often not very credible. Thus, with this work, we intend to create reliable sources of knowledge to be used both by researchers for the formulation of new theories and discovery of new therapies related to gluten, and by patients who are just looking for answers to their health issues. With the methodologies applied, we were then able to create models to reduce the time and cost required to curate documents related to the gluten bibliome.