The purpose of this study is to develop and evaluate a system that annotates health information ontology to health information contents to improve search of the National Health Information Portal, develop a health information ontology utilizing the sy...
The purpose of this study is to develop and evaluate a system that annotates health information ontology to health information contents to improve search of the National Health Information Portal, develop a health information ontology utilizing the system, and then annotate health information ontology concept to health information contents and evaluate the suitability of annotations.
This study was categorized into three stages. The first step was to develop a system that creates health information ontology that connects back to health information content. In the second step, a health information ontology was developed using the system. In the last step, health information ontology concepts were annotated to health information contents.
The health information ontology development and health information content annotation system received heuristic evaluation and PSSUQ evaluation from five experts. Heuristic evaluation disclosed two items with a severity level of three or higher that must be corrected due to usability problems. As a result of the PSSUQ evaluation, the average score for each item was 3.4 to 5.0 points, with the overall average score of 4.3 points. The health information ontology was developed with 3,823 concepts and 12,082 synonyms. The health information ontology was evaluated as having no errors in its structure and received an average of 4.4 points in content evaluation. Also, the evaluation confirmed that no structural error was indicated in the health information ontology, resulting in an average of 4.4 points. As a result, 14,740 ontology concepts were annotated to 3,241 contents. In the expert evaluation on annotation suitability, three concepts out of fifty concepts were evaluated as inappropriate.
The significance of the health information ontology development and health information content annotation system developed in this study is to automatically annotate ontology concepts to the health information contents, whereas previous ontology development and management system was limited to ontology itself, which cannot be directly linked to the information contents utilizing available ontology information. The developed health information ontology received positive evaluations in both structure and content. The suitability of health information content annotation was created based on the principles of health information ontology and health information content annotation. Some results were improved according to the results evaluated by experts.
In the future, integration between the ontology development and health information content annotation system with CMS is vital in order to apply to the content development process. In addition, it is necessary to automate the collection of user terms so that the ontology-based search function can be strengthened. In addition to providing ontology-based search, generating personalized health information contents by exploiting user’s search behavior is essential.