In the context of rapid development in the information age,the widespread application of big data technology has penetrated into various industries,bringing massive data resources.However,how to extract useful information from these data and provide support for enterprise decision-making has become an urgent problem to be solved.This study aims to optimize the data mining algorithms of big data platforms and improve the accuracy of power data mining through system design and technical implementation.Adopting algorithmic optimization methods including model compression,parameter tuning,and parallel computing to improve the performance of deep learning models.Based on this,an efficient and scalable data mining platform has been constructed.After optimization,the deep learning model has a prediction accuracy of over 95％.In addition,through parallel computing and distributed storage,the processing speed of data mining platforms has been doubled,enabling them to process larger datasets.The application of optimization methods has significantly improved the performance of the model and the processing ability of the platform,providing technical support for big data mining.
big data platformdeep learningdata mining algorithms
NETL
NSTL
万方数据
