Identify Appropriate Data Attributes in order to Knowledge Discovery from Health Big Data Systems

This study aimed to introduce the factors confirming the desirability of the data attributes in order to knowledge discovery from health big data systems by using factor analysis.

This research was a Survey-Analytical study. The statistical population consisted of 655 specialists in the fields of data science, bioinformatics, and health information analysis inside and outside Iran, who were identified by snow flack method, and the research tool, researcher-made questionnaire was sent for them, and 247 questionnaires were completed and turned back. Data analysis methods were exploratory and confirmatory factor analysis using SPSS 22.0 and Lisrel 8.8 software.

The results of Exploratory Factor Analysis explained six factors for identifying data characteristics that are: data quality, data stability, data compatibility, data value and accessibility, data efficiency and accuracy and completeness, which altogether had 68% of the total variance of the data. Confirmation analysis fit indices showed that the six-factor model has an acceptable fit with the data.

Findings indicate that based on the proposed model, data properties affect the process of knowledge discovery of health big data systems and six identified factors and 20 items of these factors, can be a good tool to identify the characteristics of data suitable for knowledge discovery of health big data systems for the effective use of these systems, emphasizing the independence of data format, and can help in the early diagnosis and treatment of diseases.