The burgeoning digitization of scientific research and the concurrent proliferation of sensitive data emphasize the pressing need to investigate and enhance research security practices among scientists. This study outlines a comprehensive knowledge discovery endeavor that leverages machine learning techniques to analyze a survey designed to uncover insights into the current state of research security practices within the scientific community. The study focuses on a survey conducted among scientists in the Faculty of Science, University of Ibadan to gain insights into their awareness, adoption, and perceptions of research security measures. It seeks to identify the prevailing trends, challenges, and gaps in security practices that may compromise the integrity and confidentiality of scientific research. Through analysis, machine learning and visualization techniques, the study uncovered valuable patterns and correlations within the survey data. The knowledge discovery process in this study involved examining factors such as researchers’ status, years of experience, knowledge of dual-use research, medium of data storage, collaboration experience, training on research security and risk identification among others. The outcomes of this research encompass the identification of common security vulnerabilities, best practices, and potential areas for improvement in safeguarding scientific research data. Hence, the results of this study is a potential tool to inform policy development, enhance security awareness initiatives, and guide the scientific community in strengthening its defenses against threats to research integrity.

Research security Dual-use technology Machine learning Best practices Bootstrapping