BIG DATA-BASED RESOURCE RECOMMENDATION METHOD AND APPARATUS, AND COMPUTER DEVICE AND STORAGE MEDIUM

摘要

A big data-based resource recommendation method, relating to data processing technology, and comprising: acquiring a training data set and a plurality of base models, and training the base models to obtain standard base models; predicting, by means of the standard base models, a second data subset comprised in the training data set to obtain at least two prediction result sets; classifying the prediction result sets by means of a classification model to obtain a first recommendation resource set; adding implicit feedback information of resource data into the first recommendation resource set to obtain a second recommendation resource set, and training a pre-constructed recommendation model by means of the second recommendation resource set to obtain a resource recommendation model; and inputting acquired target resource data and/or target user data into the resource recommendation model, generating a resource recommendation list, and performing resource pushing. In addition, the target resource data and/or the target user data can be stored in a blockchain node. The method can solve the problems of low resource recommendation efficiency and high costs.