Association Rules To Identify Receptor And Ligand Structures Through Named Entities Recognition

One of the challenges in natural language processing (NLP) is to semantically treat documents. Such process is tailored to specific domains, where bioinformatics appears as a promising interest area. We focus this work on the rational drug design process, in trying to help the identification of new target proteins (receptors) mid drug candidate compounds (ligands) in scientific documents. Our approach is to handle such structures as named entities (NE) in time text. We propose the recognition of these NE: by analyzing their context. In doing so. considering; an annotated corpus on the RDD domain, we present models generated by association rules mining that indicate which terms relevant to the context point out the presence of a receptor or ligand in a sentence.