HealthTranslator: automatic annotation of Web documents in order to assist health consumer's searches

• Autor(a) principal: Hugo Miguel Ribeiro de Sousa
• Data de Publicação: 2016
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://repositorio-aberto.up.pt/handle/10216/85756
• topic: Engenharia electrotécnica, electrónica e informática; Electrical engineering, Electronic engineering, Information engineering

The Web is now one of the main sources of health related information. However, health consumers do not always easily understand the retrieved information, mainly because of a significant gap between terminologies used by laypeople and medical experts. This work presents a tool, available as Google Chrome extension, that helps users to overcome the difficulties they face when reading health Web documents. It provides automatic annotation of medical concepts in Web documents with additional information, such as concept definition, relationships with related concepts or linkage to external resources. Both Portuguese and English languages will be supported. In order to evaluate the developed solution, its' annotation will be compared with Portuguese and an English corpus. While the first one will be manually annotated, the latter is an automatic annotation performed by a similar extension, named Medical Translator. A user study will also be conducted, in order to understand their opinion and evaluate the tool utility. It is also presented the planning for the dissertation, in order to achieve the defined goals.