Evidence-based dietary information represented as unstructured text is a crucial information that needs to be accessed in order to help dietitians follow the new knowledge that arrives daily with newly published scientific reports. Different named-entity recognition (NER) methods have been introduced, but all of them require annotated corpora from desired domain of interest. Creating annotated corpora is a demanding, time-consuming job that can take years.

JSI has developed a novel rule-based NER method, for information extraction of evidence-based dietary recommendations. Evaluation of the method showed that the method gives promising results and can be used for information extraction of evidence-based dietary recommendations.