Mutual Information Mining for Component Law and Development of New Recipes of Topical Herbs for Atopic Dermatitis
Traditional Chinese medicines (CM) topical herbs for atopic dermatitis are mainly consist of nourishing the blood, dryness-moistening, dry-dampness, skin-moisturizing and itching-relief, of which therapeutic principles are taken as dryness-moistening to relieve itching. Based on the modified mutual information, complex system entropy cluster and unsupervised hierarchical clustering, the CM recipes for atopic dermatitis can be better collected and stored in the database, and the correlation coefficient between herbs, core combinations of herbs and new recipes also can be analyzed. In our study, the objective is to evaluate and analyze the component law of CM topical herbs and explore new recipes for atopic dermatitis through data mining methods so as to provide references for dermatologists in the clinical practice and decision-making for the treatment of atopic dermatitis.
The core combinations of Chinese herbs is attained by hierarchical clustering, and the core combinations occur in the original database and occur in a certain compound recipe, however, it could not be considered as a new recipe, therefore, the hierarchical clustering carried out necessarily. It is based on a hypothesis: A new recipe is the recombination of a strong correlative core combination. Hierarchical clustering and correlation calculation about are necessary, both involved in calculating the correlation between classes. To ensure that the combination attained by hierarchical clustering is a new recipe, i.e., it does not exist in original database, the mutual information needs to be improved. Such can assure the present recipe is a new one.
Here, pro(X,Y)>0 expresses the frequency of core combination X and Y occurrence. The frequency>0 means that it occurs in a certain compound prescription, so it is not a new recipe. Thus, it is necessary to decrease the correlativity, making the correlativity of the core combination in the new recipe reach to the maximum. H(X)and H (Y) represent information entropy of X and Y respectively, while H(X,Y) expresses combination entropy[5,
Since 2010, some specialist used the unsupervised date mining methods to explore the component law and recipes of Chinese medicine for H1N1, the date mining methods has start to expand gradually, thus, we can explore the treatment of AD of Chinese medicine by this methods conveniently. Based on this method, the quintessence of ancient famous Chinese medicine specialist and famous ancient recipes can be effectively inherited. Moreover, the accuracy of screening herbs can be ensured and the limitation of personal experiences also can be better avoided. Therefore, this is a convenient tool for research of Chinese herbs. Based on the design above-mentioned, we can conclude the result of frequency analysis, core combinations and the new recipes, the CM topical herbs for atopic dermatitis that mainly consist of blood-nourishing, dryness-moistening, dry-dampness and relieve-itching also can be effectively evaluated and figured out. Moreover, by analyzing the internal relations among the commonly-used drug pairs, the function of drug Pairs may be made sure such as opening the orifices with aroma, activating blood, moisturizing the skin. The results might not be entirely consistent with the TCM clinical practice, of which possible cause lies as follows: the number of recipe prescriptions is not enough. Therefore, the research can provide references for dermatologists in the clinical practice and decision-making for the treatment of atopic dermatitis. Know More