User portrait scoring method and device and storage medium

Abstract

The invention provides a user portrait scoring method, a user portrait scoring device and a storage medium, which are used for improving the accuracy and timeliness of user portrait evaluation. The user portrait scoring method comprises the steps that a current behavior data sequence of a target user is collected, and the current behavior data sequence comprises a plurality of behavior data; basedon the current behavior data sequence, a behavior classifier set obtained through pre-training is used for determining an abnormal score corresponding to the target user, the behavior classifier setcomprises a plurality of behavior classifiers, and each behavior classifier is obtained through training by using the historical single-class behavior data sequence of the target user; searching a first hidden state corresponding to the current behavior data sequence according to an observation state corresponding to the current behavior data sequence; and determining a portrait score corresponding to the target user according to the transition probability from the first hidden state to the second hidden state and the abnormal score.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a user portrait scoring method, device and storage medium. Background technique [0002] User portrait is a way to describe users and connect user needs and product design. It aims to extract a user's information as comprehensively and meticulously as possible from massive user behavior data, so that the collected data has practical significance. and value. With the continuous expansion of application fields, the concept of user portraits is gradually changing. Generally speaking, user profiling is the labeling of user information, which is a technology that comprehensively mines a series of content such as user personal information, social information, and behavior information, and realizes the classification of different users. Its core is to represent and store the user's potential intentions and interests, and to summarize a readable and computable u...

AI Technical Summary

Problems solved by technology

However, because the template rule method relies on manual definition and cannot dynamically follow user behavior, it will cause false positives and false positives for abnormal behaviors, reducing the accuracy of portrait evaluation. On the other hand, the existing evaluation methods cannot be fast enough. The detection and inspection of abnormal personnel and abnormally used accounts affect the timeliness of portrait evaluation

Images