Analysis And Application Research Of Recommendation Algorithm Based On Collaborative Platform

With the rapid development of computer technology,people can already apply many knowledge in the computer field to solve many daily problems,and can abstract the problem in the form of modeling to solve one or more types of problems,Be a true bypass.However,the collection of large amounts of data makes a lot of potential information that has not been noticed,but such a huge amount of data also causes many problems,such as information overload and waste of resources.These problems have caused the trouble of not getting the required information accurately.In order to better solve such probl ems,personalized recommendation systems came into being.The recommendation system collects and analyzes a large amount of user information to understand the user's likes and dislikes,and then considers the characteristics of its classification and induction,and then forms recommendations,so that users can more easily achieve information acquisition and utilization.First of all,this article gives a detailed introduction to the development environment and development status of the recommendation algorithm,and explains the overall background of the current research.Secondly,it describes and implements the existing classic algorithms,clarifies the working principle of the recommendation system,classifies the recommendation algorithms,shows that different recommendation algorithms have significantly different recommendation emphasis,and they also have unavoidable defects.Based on this idea,a data pre-processing method is proposed to optimize the defects of the classic algorithm,and two scoring pre-processing algorithm optimizations of the parallel relationship are proposed.Thirdly,after in-depth study of the classic Slope One recommendation algorithm,the Slope One recommendation algorithm based on association rule strategy weighting is proposed.The algorithm can improve the sparsity of user rating data and deal with the problem of poor correlation between items,and get recommendation results that are more accurate and more in line with user needs.From the analysis of the existing recommendation system application,in the field of machine learning,the recommendation idea is reflected in many algorithm directions.For example,in the field of data mining,re-analysis of data classification requires a good outlet to use the data more fully Key info rmation presented.The single-use recommendation algorithm has not yet used the resources to the maximum,and can be more applied to the construction of smart life through transfer learning,making the related areas of smart life concept better The coordination of the world has made the concept into reality,truly realized the wisdom concept,and pushed the development of human society to a higher level.