Recall model training method and device, recall model recommendation method and device and electronic equipment

The recall model training method and device, the recommendation method and device and the electronic equipment are applied to the technical field of information, and the method comprises the steps that at least one search word sample and the feature similarity of a plurality of first recall results corresponding to each search word sample are obtained; obtaining first selection information for each first recall result in a first preset historical time range, and selecting one or more first recall results from the plurality of first recall results as first negative samples according to the first selection information; at least one search word sample and one or more first negative samples are input into a to-be-trained recall model for calculation to obtain a corresponding calculation result, training is carried out according to the calculation result until a preset iteration stopping condition is met or the current loss is smaller than a preset loss threshold value, a trained recall model is obtained, and the