This study  describes a machine learning approach to reveal the rules for optimal treatment to improve Type 2 Diabetes (T2D) risk parameters per individual. The  results suggest a promising recommendation system to support physicians in giving personalized medicine advice to their patients.  The system is based on combining of several decision tree classifiers and based on data collected from nine Random Clinical Trail studies   of different  intermittent fasting  interventions in human.  The ability to predict whether a specific intermittent fasting intervention is successful in improving T2D risk parameters is already achieved and published   in a previous study of mine A Machine Learning Approach to select the type of Intermittent Fasting in order to improve health by effects on Type 2 Diabetes . Bioinformatics 2020. However, the results of the new recommendation system presented in this study enable us to rank the different intermittent fasting methods and choose the most efficient method for the recommendation.

Shula Shazman received her B.Sc. in Computer Sciences from the Technion, Israel in 1993. In 2011 she has finished her M.Sc. + Ph.D. in Biological Sciences, In the faculty of biology, at the Technion, Israel. The title of thesis is 'Computational approaches for characterizing Protein-Nucleic-Acid Binding'. During 2011–2015 she has been a postdoctoral fellow in the Department of Biochemistry & Molecular Biophysics, Columbia University, New-York, USA. Currently Shula works at the Department of Mathematics and Computer Science, The Open University of Israel. Their current projects are 'Disordered proteins' and 'Intermittent Fasting as a tool to treat Type 2 Diabetes'.