Research On Formal Concept Analysis Method Based On Fuzzy Relation

Formal concept analysis is an effective tool for data analysis and rule extraction.It has been widely used in machine learning,data mining,software engineering and other fileds.However,the traditional formal concept analysis theory can not deal with fuzzy information.With the emergence of massive data and the uncertainty of information,how to deal with these fuzzy information with formal concept analysis theory has become an important research content.In recent years,personalized recommendation systems have gotten a lot of attention and research.The development of collaborative filtering recommendation system has further promoted the use of application of personalized recommendation system widely.Effectively improving accuracy of collaborative filtering recommendation algorithm in dealing with fuzzy data,sparse data and incomplete data,will has a great significance for improving the performance of the recommended system.Based on the incomplete fuzzy formal context,this paper studies the basic theory of formal concept analysis and rule extraction algorithm and applies it to collaborative filtering recommendation algorithm,and applies this algorithm to collaborative filtering recommendation algorithm.The main achievements of the paper are as follows:1.We define the approximate fuzzy concepts and compatible fuzzy concepts,establish an approximate fuzzy concept lattice,identify apartial ordering relationship between compatible fuzzy concepts in an incomplete fuzzy formal context,and design the association rules mining algorithm.The experimental results show the effectiveness of the algorithm.2.Determining the sparseness of the data set by analysing the target user behavior records,we put forward some of the method of filling for the sparse data set,then gather those users of similar to the one concept using the extracted association rules and fuzzy concepts,finally find the nearest neighbor similarly with the target user,and recommend them for the user.Aiming at the incompleteness of data,a method of collaborative filtering with compatible fuzzy concept is also proposed.In summary,this paper studies the formal concept analysis method based on incomplete fuzzy formal context,and applies it to collaborative filtering recommendation system,which effectively solves the problems of data deletion,sparse and data blur.But the algorithm also has some limitations,how to improve the recommendation efficiency,solve the collaborative filtering cold start and scalable problem are the future research work.