| Internet +,Big Data,Machine Learning are today's advanced techniques,and the development trend of various industries.Fitness industry is also facing a huge reform needs in the new wave of technology.The traditional fitness mode has some flaws,such as the closeness of the information,the difficulty of accessing fitness information and user's exercise data is not fully utilized.With the rapid development of Internet +,Big Data and Machine Learning,people's demands of accessing information is increasing,the traditional fitness mode cannot meet the development needs of contemporary society.To deal with the problems,in this paper,we use the Internet +,Big Data and Machine Learning techniques to design a new kind of smart & green fitness system.This new fitness system can collect fitness data and then upload fitness data to the cloud server.Users can view their own exercise data and ranking information through mobile phone APP whenever and wherever possible.While focusing on the problem of data explosion when Internet + is fully integrated with fitness industry,we set up a Big Data platform to deal with massive fitness data and use Machine Learning algorithm to analyze the collected exercise data.The main work of this paper is divided into the following three parts:Building a new fitness system based on Internet +.This paper creatively proposes a method of reading user's fitness data from the fitness equipment which is equipped with a NFC chip by using NFC-enabled smartphones.Through independent development of mobile fitness APP and back-end server system,we can finally collect,upload and view fitness data.Moreover,we use WebSocket technology to achieve real-time data updates.Building Spark-based Big Data processing platform.This paper chooses Spark as our platform by comparing current popular Big Data platforms.Then using multiple computers to set up Spark cluster and using cluster's multi-node computing ability to analyze and calculate fitness data on the cluster.Building an intelligent recommendation system of fitness equipment based on Big Data platform.This paper uses mobile fitness APP to obtain users' equipment rating data and uses collaborative filtering recommendation algorithm to analyze users' rating data on the Spark cluster.Finally realize the function of fitness equipment intelligent recommendation.This paper is the first time proposes the method of combining Internet +,Big Data and Machine Learning with traditional fitness industry.This is a positive exploration of the future development model of the fitness industry. |
