Transaction two-party relationship information identification method and device

Abstract

The embodiment of the invention provides a transaction two-party relationship information identification method and device. The method comprises the steps of obtaining a knowledge graph constructed based on multiple relationships among multiple users; wherein the knowledge graph comprises a plurality of nodes corresponding to a plurality of users and a plurality of categories of connection edges constructed corresponding to a plurality of relationships, and the plurality of users comprise natural person users and / or enterprise users; performing graph embedding processing on the knowledge graphby utilizing a pre-trained graph embedding model to obtain a node embedding vector corresponding to each node; respectively forming triples by the two node embedding vectors corresponding to the twousers involved in the target transaction and the edge embedding vectors of the connection edges of each category, inputting the triples into a pre-trained prediction model, and predicting evaluation scores corresponding to the triples through the prediction model; identifying the relationship category information between the users based on the evaluation score so as to assist in determining whether the transaction is safe or not.

Description

technical field [0001] One or more embodiments of this specification relate to the field of machine learning, and in particular to a method and device for identifying relationship information between transaction parties. Background technique [0002] The third-party payment platform has improved the convenience of people's daily operations such as transfer and payment. In order to ensure the security of transactions, its risk control system will intelligently identify the possible risks of transactions, especially for those involving large-value transfers. Transactions between enterprise users have always adopted stricter risk control strategies. [0003] Some current risk control schemes evaluate the possible risks from the perspectives of user behavior, hardware equipment used in transfers, and network environment. Although such schemes can improve transaction security to a certain extent, they ignore other factors that can reflect Some key data or information for transac...

AI Technical Summary

Problems solved by technology

For example, when there is some commonsense information that can indicate that the probability of risk in the transaction is very low, some schemes under the existing technology still adopt relatively strict control strategies because they cannot recognize the commonsense information, so that users Still need to cooperate with the cumbersome verification operation, which affects the user experience

Images