Intelligent Question Answering System For TCM Efficacy

Intelligent question answering system is a hot topic in the field of artificial intelligence and plays an increasingly important role in human daily life.However,due to the complexity of natural language,it is very difficult for computers to correctly understand human language,especially in some special fields with a large number of professional terms,the application effect of intelligent question answering system is still far from meeting the needs of users.The purpose of this paper was to integrate the intelligent question answering technology with the field of traditional Chinese medicine(TCM),build an intelligent question answering system for the field of TCM,and satisfy people's thirst for knowledge related to TCM.However,the text description form of traditional Chinese medicine is very flexible and changeable,as well as its unique expression mode of "vernacular Chinese + classical Chinese",which make the natural language processing technology can not be applied well,which increases the difficulty of constructing the question answering system.Therefore,in the process of building the system,the following researches were made:1)Data collection and collationIn this paper,before the construction of the system,the existing database of traditional Chinese medicine and related websites were collected and sorted out.Nearly 15,000 pieces of data of Chinese herbal medicine,8,000 pieces of data of Chinese patent medicine,and nearly 25,000 pieces of data of Chinese medicine prescription were integrated,which were stored in the database as data support for the construction of the system.2)Research on word segmentation method based on TCM textsIn view of the unique expression of Chinese medicine sentence,this paper introduced the named entity recognition network model based on Chinese medicine in the process of constructing the system to predict the entities with important expression significance in the text.At the same time,the concept of external dictionary was introduced,and the idea of "prefix dictionary" and "dynamic programming" was usedto complete the segmentation of TCM texts.3)Research on the similarity calculation method of TCM sentencesAccording to the particularity of Chinese medicine sentences,this paper firstly used the similarity calculation method that combined string matching and semantic analysis for words with important expressions in sentences.Combined with the rank list of constructed words,the weights of each rank were dynamically assigned,and the similarity between sentences was calculated based on the distribution results.4)Development of an intelligent question answering system based on We Chat appletThe function of We Chat applet was designed based on the collected and sorted data of TCM,and the natural language processing method was used to realize each function module,thus a intelligent question answering system capable of effective human-computer interaction was built.