Publication: Aplicación de técnicas de aprendizaje automático para la extracción de información en textos farmacológicos
Loading...
Identifiers
Publication date
2011-07
Defense date
2011-07-15
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
En la actualidad los profesionales del dominio biomédico necesitan tener información actualizada de su campo para llevar a cabo su trabajo de manera fiable y profesional. Dentro del dominio biomédico, la administración de fármacos requiere saber de antemano si dos fármacos interaccionan entre sí, ya que esta interacción puede provocar efectos no deseados en la salud del paciente. Los profesionales cuentan con ingentes cantidades de información, ya sea en textos biomédicos no estructurados o en bases de datos; es por esto que se necesita un método automático para extraer información de estas fuentes de datos para poder detectar interacciones entre fármacos. En este proyecto se van a estudiar distintas técnicas de aprendizaje automático supervisado para detectar posibles interacciones entre dos fármacos. Partiendo del corpus DrugDDI, creado en la tesis Application of Information Extraction techniques to pharmacological domain: Extracting drug-drug interactions, se van a aplicar diferentes algoritmos para su posterior estudio y comparación con los resultados obtenidos en dicha tesis. ____________________________________________________________________________________________________________________________
In the biomedical domain, interaction between two or more drugs is a desired knowing in drugs administration, as that interaction can provoke undesirable effects over a patient health. Medical professional have access to huge amounts of data, whether they are in biomedical unstructured texts or in databases. For this reason it is desirable an automatic method to extract useful information from this data sources for processing and detecting drugs interactions. In this project we are going to introduce some supervised machine learning techniques in order to detect possible interactions between two drugs. Based on the DrugDDI corpus, gathered in the thesis Application of Information Extraction techniques to pharmacological domain: Extracting drug-drug interactions, we are going to apply different algorithms for its later research and comparison with the results obtained in the thesis.
In the biomedical domain, interaction between two or more drugs is a desired knowing in drugs administration, as that interaction can provoke undesirable effects over a patient health. Medical professional have access to huge amounts of data, whether they are in biomedical unstructured texts or in databases. For this reason it is desirable an automatic method to extract useful information from this data sources for processing and detecting drugs interactions. In this project we are going to introduce some supervised machine learning techniques in order to detect possible interactions between two drugs. Based on the DrugDDI corpus, gathered in the thesis Application of Information Extraction techniques to pharmacological domain: Extracting drug-drug interactions, we are going to apply different algorithms for its later research and comparison with the results obtained in the thesis.
Description
Keywords
Ingeniería del conocimiento, Aprendizaje automático, Recuperación de la información