Research On Technology Of Recommending Experts For Innovative Design Of Mechatronic Products Under Social Network Environment

Under the influence of the globalized market economy,the competition for mechatronic products has become increasingly fierce.The implementation of innovative design is an effective mean to enhance the competitive advantages of the product market.Innovative design is a kind of design activity with low cost and high efficiency,which integrates multiple fields of knowledge and experience,and how to find suitable innovative design experts is one of the points.Nowadays people often communicate through the Internet,but there is information overload phenomenon in the network.How to find out the domain experts who are interested in and trust in massive information is a difficult problem to face.Recommended technology is an effective way to solve the problem of information overload,and it has been successfully applied to various areas of the Internet.However,the traditional recommendation technology has problems such as low recommendation accuracy and insufficient coverage.With the development of Web 2.0 technology,social networks provide a new convenient platform for people to communicate and collaborate.The effective combination of social network theory and recommendation technology can make up for the above shortcomings of traditional recommendation technology.In this paper,the key issues of recommendation technology and expert recommendation are studied and explored around the user and text information in social network.First,the three basic elements of the recommendation expert: search,recommendation,and ranking are studied,and a LDA topic model based on text semantics is established to fully excavate the potential semantic topics of the text content.Then,we analyze the social relations between users in the social network,the trust relationship between users is introduced into the recommendation process of experts,to construct a method based on users trusted LDA subject model domain experts recommendation method.And the feasibility and effectiveness of the algorithm are verified by experiments.Finally,use Java language to write programs,development experts recommend prototype system.Through the prototype system,we use the microblogging information in the social network to discover the experimental test of innovation design experts in mechatronic products.This method is based on the LDA theme model,the text content related to specific users in Sina micro-blog is matched with the theme of innovation design in mechatronic products through semantic similarity.Considering the social network structure and the trust relationship among users,a trust network between users is constructed.The user who has a high degree of similarity to micro-blog content semantic topic and is trusted by the user is sorted,the top 20 users are returned to the user as the final recommendation experts.The method can make up for the traditional expert recommendation method that only considers the characteristics of expert knowledge and expert network,to ignore social relationships between users,which leads to the low accuracy of expert recommendation and the standardization of recommendation results.