Dish recommendation method and device

Abstract

The invention discloses a vegetable recommendation method and a vegetable recommendation device, wherein, the method comprises the following steps: user attribute information and scene attribute information of a target user are mapped into corresponding attribute information vectors; Acquiring a vegetable vector of a plurality of dishes to be recommended; The attribute information vector and the vegetable vector of each dish to be recommended are inputted into the correlation degree model obtained by training, and the correlation degree value of the attribute information vector and the vegetable vector of each dish to be recommended is obtained; The recommended dishes are selected from a plurality of dishes to be recommended according to the correlation value and recommended to the targetuser. In accordance with that invention, dishes can be recommended without relying on the historical ordering behavior of the target user, Even for the target users whose ordering behavior is not rich, the relevance value between the attribute information vector of the corresponding target users and the vegetable vector of the food to be recommended can be obtained by using the relevance model, and the ordering experience of the target users can be improved by using the relevance value.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a dish recommendation method and device. Background technique [0002] With the rise of smart catering technology, the way of ordering food has evolved from traditional paper ordering to terminal intelligent ordering, for example, ordering food on a takeaway platform or ordering food in an ordering terminal provided by a restaurant. At the same time, in order to increase the order rate of dishes and enhance the user's ordering experience, more and more platforms that provide ordering services will recommend dishes to users through terminals. [0003] In the existing recommendation schemes, the user's preference information is usually determined according to the user's historical data, and recommendations are made based on the preference information. For example, in the Chinese patent application with application publication number CN1068157745A, the target user's dish componen...

AI Technical Summary

Problems solved by technology

This scheme relies on the target user's historical interaction behavior to determine the target user's preference for the target object, and cannot accurately predict the preference degree of the target user who has no or less interaction with the target object, and thus cannot make accurate recommendations

Images