Big data prediction method for positioning shop where user is located based on partial mark learning

Abstract

The invention requests to protect a method for predicting positioning big data of a shop where a user is located based on partial mark learning. The method comprises the following steps: 101, carryingout preprocessing operation on shopping state data of the user; 102, constructing a partial mark data set according to the candidate shop set corresponding to each sample; 103, performing feature extraction operation on the offset mark data set; 104, constructing a similarity graph according to the feature space; 105, performing probability propagation according to the similarity graph; 106, predicting the shops with behavior interaction in the future of the user from the candidate shop set of the partial mark data set through the convergence probability of propagation. The method mainly comprises the following steps of preprocessing user historical data, extracting features, converting the features into partial mark data sets, establishing a partial mark learning model, and predicting shops with behavior interaction in the future from candidate shop sets corresponding to all users according to the partial mark data sets of the position behaviors of the users, so that the users can obtain more accurate personalized push services, and the shopping experience of the users is improved.

Description

technical field [0001] The invention belongs to the technical field of partial label learning and big data processing, and in particular is based on the prediction of big data of store location where a user is located based on a probability propagation model. Background technique [0002] Partial label learning is a weakly supervised learning in which the output space is associated with a set of candidate labels. Only one of the candidate label sets is the real label, and the remaining labels are regarded as interference noise labels. In the process of partial label training, the real label of each training sample is submerged in the candidate label set, so it is impossible to directly obtain the learning algorithm from the input space to the output space from the data set similar to the strong supervised learning. However, in real life, datasets with accurate and unique label information are more and more difficult to obtain. So we have to face the serious problem of how t...

AI Technical Summary

Problems solved by technology

In the process of partial label training, the real label of each training sample is submerged in the candidate label set, so it is impossible to obtain the learning algorithm from the input space to the output space directly from the data set similar to the strong supervised learning.

Images