Research On Individual Portrait Based On User Value Of Home Appliance Scene

Production-oriented manufacturing enterprises need to transform into service-oriented manufacturing enterprises due to low added value of products and single form of profit.As one of the traditional production-oriented manufacturing enterprises,in the process of transformation to service-oriented manufacturing,home appliance enterprises need to first grasp the personalized needs of users and the operation status of equipment,and then provide personalized services to users and realize value-added services.Therefore,this thesis takes the home appliance scenario as the main research object,and constructs individual user portraits and individual device portraits from two perspectives: individual user and individual device,combining device usage data and device operation data,so as to help enterprises understand individual user’s functional preferences and usage habits,and grasp the health status information of devices in the region,thus helping enterprises provide personalized services to users and realize service value-added services.The main research work of this paper is divided into the following three parts.(1)Research on individual user profile method.In this paper,the FP-growth algorithm is optimized by cutting invalid branches to reduce the running time of the algorithm and the space resources consumed by the algorithm.After that,the improved FP-growth algorithm is combined with device usage data to build an individual user portrait model to obtain information on individual users’ device function preferences and device usage habits,and the experimental results show that the improved FP-growth algorithm has better mining effects and can better help enterprises grasp users’ function preferences and usage habits,so as to provide personalized services to users.(2)Research on individual portraits of device health status.The optimization direction of the particle swarm algorithm(PSO)is first targeted to improve the optimization performance of the algorithm by using the Corsi variance and dynamic inertia factors.After that,the optimized particle swarm algorithm(IPSO)is used to optimize the regularization coefficients and function width of the least squares support vector machine model(LSSVM)to improve the classification performance of the LSSVM model,and then the IPSO-LSSVM-based individual portrait model of equipment health status is constructed.The experimental results show that the IPSO-LSSVM model is more accurate than the PSO-LSSVM model and the standard LSSVM model in determining the health status of equipment,and can help enterprises grasp the health status of equipment in the region more accurately.(3)Research on individual portrait method of equipment life.Firstly,the optimization direction of Sparrow Search Algorithm(SSA)is targeted to improve the optimization performance of the algorithm by using a hybrid strategy based on good point set,golden sine strategy and adaptive t-distribution variation strategy.After that,the optimized sparrow search algorithm(ISSA)is used to optimize the parameters of the data center,network weights and kernel function width of the RBF neural network model to improve the prediction performance of RBF,and then the ISSA-RBF-based individual portrait model of the remaining life of the device is constructed.The experimental results show that the prediction effect of ISSA-RBF model is better than the standard RBF model,which can help enterprises grasp the remaining working life of the equipment in the region.Finally,the thesis constructs an individual portrait system based on the user value of home appliance scenarios and applies the above model to realize the functions of individual user portrait,individual portrait of equipment health status and individual portrait of remaining equipment life,so as to provide a basis for personalized services and operation and maintenance services for enterprises and thus realize value-added services.