E-commerce session recommendation method, system and device based on multi-behavior feature fusion and medium

The invention discloses an e-commerce session recommendation method and system based on multi-behavior feature fusion. The method comprises the following steps of: obtaining four session behavior data of clicking, collecting, purchasing and adding shopping carts of a user on commodities in an e-commerce database; an electronic commerce session recommendation model based on multi-behavior feature fusion is constructed, and the model comprises a behavior feature extraction module, a commodity feature extraction module, a commodity high-order feature extraction module, a session feature extraction module and a commodity recommendation module; four kinds of session behavior data in an e-commerce database are used for training the e-commerce session recommendation model; and utilizing the trained e-commerce session recommendation model to carry out commodity recommendation on users in the session, and outputting a recommendation result. The multi-behavior characteristics in the session are fused by using the e-com