Information extraction from binary skill assessment data with machine learning

Abstract

Strength training exercises are essential for rehabilitation, improving our health as well as in sports. For optimal and safe training, educators and trainers in the industry should comprehend exercise form or technique. Currently, there is a lack of tools measuring in-depth skills of strength training experts. In this study, we investigate how data mining methods can be used to identify novel and useful skill patterns from a binary multiple choice questionnaire test designed to measure the knowledge level of strength training experts. A skill test assessing exercise technique expertise and comprehension was answered by 507 fitness professionals with varying backgrounds. A triangulated approach of clustering and non-negative matrix factorization (NMF) was used to discover skill patterns among participants and patterns in test questions. Four distinct participant subgroups were identified in data with clustering and further question patterns with NMF. The results can be used to, for example, identify missing skills and knowledge in participants and subgroups of participants and form general and personalized or background specific guidelines for future education. In addition, the test can be optimized based on, for example, if some questions can be answered correct even without the required skill or if they seem to be measuring overlapping skills. Finally, this approach can be utilized with other multiple choice test data in future educational research.