Information-based commodity recommendation method

The invention relates to an information-based commodity recommendation method, which comprises the following steps of: acquiring historical data, and performing early-stage labeling on the historical data by adopting a related grade labeling mode; combining the labeled grade data as a training target, and obtaining a plurality of initial recommendation models through model training; based on the score matrix, respectively calculating loss functions corresponding to the plurality of initial recommendation models, and screening out the initial recommendation model with the minimum loss function value as a recommendation model for verification; performing corresponding verification on the recommendation model for verification to obtain a final recommendation model; and inputting the actual information name and all commodity data into the final recommendation model, and outputting to obtain a corresponding recommended commodity. Compared with the prior art, the method can effectively improve the labeling efficien