Method for extracting embedding features of stores in external selling scene

Abstract

The invention belongs to the technical field of computer application, and discloses a method for extracting embedding features of stores in an external selling scene, a computer and a computer program. The method comprises the steps of extracting consumption behavior sequences of users; classifying the consumption behavior sequences of the users according to average consumption price of the storesand the types of the stores; for the classified consumption sequences, extracting training samples; building a negative sampling-based skip-gram model, and performing training by utilizing a tensorflow framework; and obtaining embedding eigenvectors of the stores. The stores are subjected to embedding feature extraction; feature information of the stores in certain aspects is obtained; for data,a one-hot feature space of a high dimension is converted to an embedding feature space of a specified dimension; and in combination with embedding features of merchants and an online model, the overall performance of an online order-placing model is improved.

Description

technical field [0001] The invention belongs to the technical field of computer applications, and in particular relates to a method for extracting store embedding features in a takeaway scene. Background technique [0002] In 2013, Google open-sourced word vector computing tool - word2vec, which attracted the attention of industry and academia. Word2vec can be efficiently trained on millions of dictionaries and hundreds of millions of data sets; the obtained training result - word embedding, is a good measure of the similarity between words. In the field of food delivery, for the real-time requirements of online business, the original dense feature and one-hot feature cannot meet the prediction delay requirement of milliseconds in time, and it is necessary to abstract the characteristics of the store as a whole; from the perspective of feature engineering , the existing feature engineering work is mainly carried out from a single dimension, and it is difficult to reflect th...

AI Technical Summary

Problems solved by technology

[0003] To sum up, the problems existing in the existing technology are: because the existing feature engineering work mainly reflects the characteristics of the store from a single dimension rather than from an overall perspective, this will lead to a flattening of the feature space; The offline is about 10 million dimensions, and the online is about 300 dimensions. The large feature quantity leads to high algorithm complexity, and it is difficult to meet the needs of online real-time computing.

Images