Enhancing Biomedical Knowledge Representation Through Knowledge Graphs

Abstract

There is a plethora of information related to the biomedicaldomain on the internet. Unfortunately, retrieving thisinformation online is challenging because of insufficientsemantic metadata embedded within the web documents thathelp search engines interpret the biomedical information.Semantic annotators have partially bridged this gap, yetthese tools frequently need to catch up in accuracy, speed,and the ability to dynamically represent knowledge. Weinitially developed "Semantically," a biomedical semanticcontent authoring platform to streamline and enhancebiomedical annotations through a social-technical approach.Even so, the current system stores data in a relationalschema, which lacks machine-readable content that allowssearch engines to parse the meanings to annotationrecommendations. There is still the need for theamalgamation and contextually rich representation ofannotation recommendation information to enhancenavigation and exploration of data. Therefore, we propose aknowledge graph-based recommendation system with annlp-enhanced search query to provide an environment foreasy and quick access to optimal recommendations in amachine-readable knowledge graph format. We obtainresults for the knowledge graph through an evaluationsurvey that substantiates the efficacy of our knowledgegraph-based recommendation system, highlighting its role inadvancing dynamic knowledge representation and semanticannotation in the biomedical domain. A demo is available atSebC750/Semantically at Knowledge_Graph_Branch(github.com)