Asian Journal Of Social Pharmacy

Abstract: Objective To establish a comprehensive evaluation model to assess plant polyphenols development projects objectively. Methods FANP model, together with pattern recognition technology was used to weight the 21 indexes of 20 plant polyphenols extract projects from 8 companies respectively. Then, a numerical model with the support vector machine (SVM) was built to evaluate all the indexes and profitability of the projects. Results and Conclusion The results show that the correction rate, the recognition rate and the rejection rate of the SVM model for plant polyphenol extraction projects are 98.2%, 97.1%, and 95.5%, respectively. Therefore, this model can be used to predict the profitability of a project, to provide reliable method for decision making, which makes the decision making process digitized and scientific.

Key words: fuzzy analysis network process (FANP), support vector machine (SVM), plant polyphenol, weight