A Multi-Objective Hyper-Heuristic Clustering Algorithm for Formulas in Traditional Chinese Medicine

Syndrome types are important for diagnosis and treatment in traditional Chinese medicine. can be summarized by domain experts as formula clusters. In this paper, we propose seven feature models the clustering problem based on categories, subcategories, functional tendencies names of materia medica. A novel multi-objective hyper-heuristic algorithm is obtained. our proposed algorithm, 12 low-level heuristics used solution perturbation merging clusters, dividing clusters or moving points between received solutions from high-level heuristic. The heuristic evaluates heuristics, updates pool, selects initial next iteration via roulette wheel selection Pareto front. Experimental results demonstrate that outperforms other algorithms most datasets. number has less influence final than algorithms. For datasets, mechanism front shows higher convergence rates accuracy a random mechanism. Accuracy was models.