Recommendation Method Ang Implementatin Of Old-age Service Based On Deep Learning And Multi-objective Optimization

With the acceleration of the aging of the country,domestic attention to the issue of old-age care has become more and more important,and healthy pension has become one of the most important issues in today's society.At present,the old-age service industry has problems such as low information integration and low data utilization.To solve the above problems,it is one of the most effective means to build a connected old-age service platform.However,due to the diversification of the old-age service resources on the platform,the elderly cannot quickly find a service that suits them,resulting in a mismatch between user needs and service resources.Therefore,it is necessary to extract the basic characteristics of the elderly and the aged care service,and to meet the individualized needs of the elderly through the way of service recommendation.To this end,this paper will build a recommendation model based on deep learning for the aged care service,generate a set of service recommendation candidates,study the multi-objective optimization recommendation algorithm based on artificial immune,and do detailed demand analysis and functional module design for the pension service recommendation system,and finally design and To implement the pension service recommendation system,the work is carried out mainly from the following four aspects:(1)Based on the cold start and data sparse problems existing in the traditional recommendation algorithm,a deep learning-based recommendation model for the recommendation of the aged care service is established.Through the basic attribute information of the elderly and the aged care service,the characteristics of the elderly and the characteristics of the aged care service are extracted respectively,and the model is trained according to the mean square error of the predicted score.Finally,the trained model is used to select the suitable target users from the many aged care services.Service,as a candidate for pension service recommendation.(2)In view of the simplification of the recommendation target existing in the current recommendation system,the user is gradually immersed in the information mortuary.The multi-objective optimization recommendation model based on artificial immune algorithm is constructed,and the recommended accuracy and novelty are takenas the objective function.The service recommendation candidate set extracted by the deep learning model is used as the initial population,the affinity between the antibodies is calculated,the dominant population is found and the cross-variation cloning operation is performed,and the final recommendation list is obtained after reaching the maximum number of iterations,thereby recommending to the elderly more accurate and updated service.(3)Analyze the functional and non-functional requirements of the pension service recommendation system,and analyze the key business processes in the pension recommendation system.Design the overall structure,function and database of the pension service recommendation system,define the data types and formats needed in the system,and design the three main aspects of the system's main function modules:class diagram,timing diagram and database entity relationship.(4)Combining the above research on the recommended algorithm and model of the aged care service,and the detailed design of the recommendation system for the aged care service based on deep learning and multi-objective optimization,the whole system is finally implemented and its performance and function are tested.