User collection-based commodity recommendation method and system

The invention discloses a user collection-based commodity recommendation method. The method comprises the steps of S1, obtaining collected commodity information of a user in a recent time period fromuser data through data extraction; S2, inquiring the selling conditions of the collected commodities in each channel in the step S1, and judging whether the collected commodities are off-shelf commodities or not at the current time; if so, executing S3; if not, directly pushing the information of the sold/preheated commodity to the user; S3, judging whether the user purchases the collected commodity within a set time range or not, and judging whether the collected commodity needs to be recommended to the user or not in combination with the category of the collected commodity; and S4, extracting the sku codes of the collected commodities, screening out the sold/preheated commodities belonging to the same type as the collected commodities according to the sku codes of the collected commodities, selecting one of the