Design And Implementation Of Refined Dynamic Information Push System For Smart Campus

With the development of information technology,significant achievements have been made in the construction of educational information.As a brand new educational environment,smart campus plays an important role in the construction of educational information.However,with the integration of more and more technologies into the smart campus,the campus has gradually become a complex area of information,which not only causes information waste,but also affects the daily life and study of teachers and students in school due to excessive information,which makes the development of smart campus step into the dilemma of rapid development of campus information and campus information redundancy.Therefore,there is an urgent problem to be solved is how to effectively use the information to meet the needs of development of campus and the daily needs of teachers and students in campus.This paper uses the advantages of information recommendation system in information processing and recommendation to design and develop a refined dynamic information push system for smart campus.By means of collaborative filtering,the information and data are provided to different users in different categories to meet the needs of users.At the same time,indoor and outdoor positioning technology is used to further determine the user's current scene,combining the user's needs with the user's environment,so that the recommendation information can better meet the user's needs.The main work is as follows:A three-dimensional positioning algorithm based on k-means clustering is improved.In order to solve the problem that the complexity of indoor positioning environment makes the positioning error large and it is difficult to realize the 3d positioning,the 3d positioning is realized while the fingerprint database is processed.The RSSI correlation coefficient is effectively integrated and weighted with the traditional algorithm to make the RSSI value at each position directly mapping the position information of the position.The validity and accuracy of the improved algorithm are verified by experiments.A collaborative filtering similarity measurement method based on weight matrix is improved.In order to make the generated recommendation information more personalized,it is considered from the two aspects of user crossover project and existing weight project besides crossover project.At the same time,the modified Tanimoto coefficient and related similarity are fused to effectively avoid errors caused by inconsistent user rating standards and improve data utilization.The accuracy and feasibility of the improved algorithm are verified by the MAE value as the standard.A refined dynamic information recommendation system for smart campus is designed and developed.Based on the improved collaborative filtering recommendation algorithm and the improved Wi-Fi fingerprint positioning algorithm,a refined dynamic information recommendation system for smart campus was designed and implemented with "Yunnan normal university" as the application background.The system is mainly divided into two modules: user location information positioning and information recommendation.The core function is to complete the refined information recommendation based on user location information,and to realize the mining of user potential information and the prediction of required information and other functions.The system test shows that the refined dynamic information recommendation system for smart campus developed in this paper can realize the information of recommendation user demand generated in the campus according to the positioning result,showing good application value.To sum up,this paper first proposed a WLAN indoor 3d positioning algorithm based on k-means clustering,which effectively improved the indoor positioning accuracy and successfully achieved 3d positioning.Secondly,a similarity measurement method of collaborative filtering based on weight matrix is proposed,which interprets the idea of collaborative filtering in different ways and effectively improves the recommendation accuracy.Finally,taking smart campus as the application environment and the proposed WLAN indoor 3d positioning algorithm and collaborative filtering recommendation algorithm as the technical support,a refined dynamic information push system suitable for smart campus is designed and developed...