Design And Implementation Of Personalized Fitness Plan Customization System

For fitness enthusiasts,they often need to collect fitness course information through gym,fitness website and other different ways.However,the amount of fitness information on the Internet makes it more and more difficult to obtain fitness courses that meet their own needs through the Internet.It is one of the main goals of fitness plan customization system that how to recommend the fitness content to users in a large number of fitness data.The application of recommendation technology is the most effective way to improve the personalized experience of fitness plan customization system.This paper designs and implements a personalized fitness plan customization system based on integrated recommendation strategy,aiming at providing personalized fitness customization system to the initial fitness users in accordance with their own conditions.Firstly,it summarizes the current development of fitness system at home and abroad,and studies and analyzes the existing recommendation algorithms.Fitness User Tag Collaborative Filtering algorithm was studied and improved,and Fitness User Tag Time Collaborative Filtering(FIT-UTTCF)was designed for the recommendation of fitness programs.Fit-uttcf makes full use of user explicit and implicit data to solve the problem of fitness plan information sparsity.By adding a time decay factor,A model that simulates user fitness preferences over time,Increase flexibility in fitness program recommendations,At the same time,a recommendation algorithmbased on Fitness User Tag Collaborative Filtering VGG16 is designed for the system optional course.The tag feature similarity and image feature similarity of fitness items were calculated,and the optimal values of recommendedmodel parameters were determined through iterative experiments.Afterthat,relevant experiments are designed to verify the good performance of the algorithm.