Research On Key Technologies Of Chinese Medicinal Materials Recommendation System Based On Multi-attribute

Nowadays,people are getting more and more information from the Internet,and recommendation systems have emerged and are widely used in all walks of life.However,there are few applications in the field of Chinese medicinal materials.In recent years,people’s attention to Chinese medicinal materials has gradually increased,purchases and related research have also increased.General Chinese medicinal materials purchase websites cannot meet the needs of people’s life and scientific research.Therefore,the paper studies the recommendation of Chinese medicinal materials based on multiple attributes.The key technology of the system,and at the same time,a prototype system for recommending Chinese medicinal materials was designed and completed.To realize the Chinese medicinal materials recommendation prototype system,firstly,the knowledge representation of Chinese medicinal materials must be completed,and the data of Chinese medicinal materials in the Pharmacopoeia of the People’s Republic of China must be converted into a language that can be recognized by the computer.Based on the full analysis of the characteristics of Chinese medicinal materials,the paper proposes an Extension-tree and uses uniform rules to describe the knowledge of Chinese medicinal materials.Constructing an Extension-tree of Chinese herbal medicine knowledge can improve the efficiency of subsequent recommendation algorithms.The paper introduces semantic similarity,fully considers the semantic relationship between Chinese medicinal materials,calculates the similarity from the three aspects of information,semantic distance and attributes,and obtains the final semantic similarity after weighted optimization by Analytic Hierarchy Process(AHP).According to the data type of the attribute value of the Chinese medicinal material,the attribute of the Chinese medicinal material is classified and calculated to make the result more scientific and accurate.The paper selects honeysuckle,Daqingye,Zhuru and other Chinese medicinal materials for experiments.The optimized semantic similarity algorithm is compared with the information-based and distance-based similarity algorithm.It is found that the basic trends of the three algorithms are the same,but the sensitivity of this study algorithm has been greatly improved,indicating that the algorithm in this study guarantees accuracy and avoids the similarity of many Chinese medicinal materials.The core of the recommendation system is the recommendation algorithm,and the paper adopts the recommendation algorithm based on the combination model.Mainly use the idea of The Content-Based Recommendation algorithm and The Collaborative Filteringalgorithm to improve the system’s prediction of user preferences while ensuring the accuracy of the recommendation results.In the calculation of the similarity of Chinese medicinal materials,the semantic similarity algorithm is introduced to improve the intelligence and semantics of the recommendation results.Increasing the threshold for the TOP-N algorithm that generates the recommendation list meets the reliability of the recommendation of special items such as traditional Chinese medicine.In this study,a prototype system for recommending Chinese medicinal materials was designed and completed.The system recommends Chinese medicinal materials to users according to user needs,and realizes user query,search,and message functions,which fully proves the feasibility of the recommendation algorithm based on the combination model.A good Chinese medicinal materials recommendation system can classify and store Chinese medicinal materials,provide convenience for people’s life and work,provide a foundation for the development of the field of Chinese medicinal materials,and carry forward my country’s traditional Chinese medicine technology.