Study on Relationship Visualization of Clinical Research-related Laws Using Word-matching

Abstract

Current clinical research is planned and pursued in accordance with the following three laws: “ GCP Ordinance,” “Clinical Trials Act,” and “Ethical Guidelines for Medical and Health Research Involving Human Subjects.” Recently, there has been an increase in studies that classify and explain laws related to clinical research.

In the domain of law information in natural language processing, calculating similarity and visualizing relationships between multiple laws is useful for managing legal costs and promoting comprehension. Similarly, it is expected that enough comprehension by simplifying and visualizing relation of laws related clinical research.

The aim of this study is to visualize the similarities between ministerial ordinances, laws, and guidelines related to clinical research. In this paper, based on the hypothesis that “the higher the degree of matching of a word, the higher the similarity between articles containing those words,” we proposed a method for measuring the relationships among possible combinations of law “articles” using “word matching,” i.e., the number of common words.

The results of performance evaluation indicated that word matching could identify the combinations of the articles with “consistence/high similar” for some items. Furthermore, high-degree word matching was observed with less meaning as a clinical research related legal word, the combinations of the articles with the “low similar/inconsistence” were acquired.