EXTRACTION OF PATTERNS USING NLP: GENETIC DEAFNESS 1 Anabel Fraga 1 , Javier Garcia 1 , Eugenio Parra 1 , Valentín Moreno 1 1 Computer Science Department, Carlos III of Madrid University Av. Universidad 30, Leganés, Madrid, Spain {afraga, eugenio.parra@kr.inf.uc3m.es, vmoreno}@inf.uc3m.es Keywords: indexing, ontologies, knowledge, genetic deafness, medicine, reuse, retrieval, conceptual text integrity. Abstract: In the domain of Genetic Deafness in medicine, it is important to detect some patterns. Medical doctors needs to search and deal with information from diverse sources and it is important to be able to cross information between sources. As part of a solution, one approach to minimize the impact of this lack and increase the success of the retrieval process crossing diverse sources of information laid in the use of Natural Language Processing techniques permitting conceptual integrity of text. 1 INTRODUCTION In this research project, we solved the following problem: is it possible to make a computer recognize linguistic patterns in a document? For example, when reading the sentence: “The patient had a temperature of 102F at Saturday” With syntactic tokens: “Determiner + Noun + Verb + Preposition + Noun + Preposition + Noun + Preposition + Noun” Is it possible for the computer to understand it both syntactically and semantically? Moreover, once this 1 DOI: 10.18293/SEKE2017-204 information is acquired, is it possible to generate patterns that allow recognizing similar sentences throughout a document? With this problem as a starting point, this project is related to the studies of Natural Language Processing (NLP). For that, by using vocabulary from a specific domain, the computer will read a number of documents and automatically generate all linguistic patterns contained in them. Therefore, there are two main objectives: • Acquire the terminology from a specific domain.