A Bootstrapping approach for investigating the consistency of assignment of plants to Jamu efficacy by PLS-DA Model

Present study investigates the main ingredient plants in jamu medicines using pls-da where the model was developed by considering plants usage in jamu as predictors and jamu efficacy as response. we utilized the coefficient matrix obtained from the pls-da model to assign plants to jamu efficacy based on the largest coefficients. however,if new jamu data set is added to the model,the coefficient configuration,and in turn the assignments,may change. thus,consistency examination of the assignments is important and bootstrapping can be used for this purpose. if a plant is useful for certain efficacy then in most bootstrap resampling rounds the plant would be assigned to that efficacy and most of the plant's coefficients corresponding to that efficacy are expected to be positive. in the present study 1000 bootstrap rounds were performed. out of 465 plants,it is found that the assignments of 276 plants are consistent and these plants are regarded as main ingredient in corresponding jamu. thus,this study gives useful information on plants serve as main ingredients in jamu efficacy by evaluating the significance of plant usage in jamu medicines using bootstrap procedure.