Research On Frequent Pattern Parallel Mining Of Large Clinical Data In Traditional Chinese Medicine Based On Cloud Platform

There is a long history and rich resources in TCM(Traditional Chinese medicine),the dissemination of academic views and ideas in the field of TCM is mainly through personal experience,practice and words of mouth between master and apprentice.This causes the development and inheritance of TCM in the contemporary society,and the speed of information processing and dissemination is slow,which is difficult to meet the needs of society for knowledge of TCM.An important engine and technical support for the development of TCM is informatization,informatization construction has been included in the national "13th Five-Year"plan.Under the support of information technology such as cloud computing technology and data mining technology,the informatization of TCM has become a hot spot of research.Attention is put on the treatment of syndrome differentiation in TCM.Clinical medicine is a basic subject in TCM,in the process of treatment based on syndrome differentiation,"knowledge intensive" data such as symptoms,syndromes,pathogenesis,treatment,diagnostic information,prescription of TCM is generated,and this kind of knowledge-based and unstructured data has become an important data for the research of clinical data of TCM.This research is devoted to the study of"knowledge intensive" data of TCM and to mine the potential useful information between"symptom-syndrome-prescription".On the basis of the existing data analysis and mining methods,the building of parallel mining framework and frequent pattern mining algorithm in cloud computing environment are mainly studied.The main innovative work of this article is as follows:1)The clinical data of TCM are diverse,the data structure is complex,and the patients with lung cancer are very different.They often show a complex syndrome of multiple diseases.The treatment is mainly composed of compound prescription,which greatly increases the difficulty of analysis.The traditional data analysis method faces challenges and difficulties.Therefore,it is necessary to take the problem of TCM as the guidance,establish a parallel cooperative mining framework for large data analysis of Chinese medicine in the cloud computing environment,and focus on the corresponding relationship among "symptom-syndrome-prescription" and the research on the core prescription.2)The classic frequent pattern mining algorithm FP-Growth algorithm mainly includes two steps:mining frequent itemsets and generating association rules.In this paper,based on the characteristics of clinical data of TCM and the research goal of mining the core prescription,the Deep FP-Growth algorithm is proposed based on the parallel collaborative mining framework.The concept of effective successor and core frequent itemsets is introduced,and the frequent itemsets with the highest correlation is excavated.3)China has a vast territory and a long history of TCM,forming a regional TCM culture,bringing incompleteness,inconsistency and abnormality of the data,and in addition to the different practice habits of different famous veteran teran doctors of TCM,which leads to the unstructured clinical data of TCM for the treatment of lung cancer.Cleaning the clinical data of Chinese medicine to treat lung cancer,which makes it a structured and standardized data.This work is particularly important.