Mining novel connections from large online digital library using biomedical ontologies

Abstract

Purpose

The huge volume of biomedical literature provides a nice opportunity and challenge to induce novel knowledge by finding some connections among logical‐related medical concepts This paper aims to propose a semantic‐based knowledge discovery system for mining novel connections from large online digital libraries.

Design/methodology/approach

The method takes advantages of the biomedical ontologies, MeSH and UMLS, as the source of semantic knowledge. A prototype system, Biomedical Semantic‐based Knowledge Discovery System (Bio‐SbKDS), is designed to uncover novel hypothesis/connections hidden in the biomedical literature. Using only the starting concept and the initial semantic relation derived from UMLS, Bio‐SbKDS can automatically generate the semantic types as category restrictions for concepts. Using the semantic types and semantic relations of the biomedical concepts, Bio‐SbKDS can identify the relevant concepts collected from Medline in terms of the semantic type and generate the novel hypothesis between these concepts based on the semantic relations.

Findings

The system successfully replicates Dr Swanson's famous discoveries: Raynaud disease/fish oil automatically, and generates much less intermediate concepts and spurious connections.

Originality/value

The method takes full advantage of the semantic knowledge of the biomedical concepts, compared with previous approaches, our methods generate much less but more relevant novel hypotheses. Another significant advantage over other traditional approaches is that our method requires much less human intervention in the discovery procedure.

Citation

Hu, X. (2005), "Mining novel connections from large online digital library using biomedical ontologies", Library Management, Vol. 26 No. 4/5, pp. 261-270. https://doi.org/10.1108/01435120510596107