Cloud edge end three-layer cooperative user behavior prediction method, system, electronic device and storage medium

By employing a three-layer collaborative user behavior prediction method involving cloud, edge, and device, the terminal device performs initial processing and transmits data to the edge device. The edge device then collaborates in processing the data, while the cloud optimizes the model. This approach addresses the issues of limited computing resources on the terminal device and insufficient computing power on the edge device, thereby achieving efficient and accurate user behavior prediction.

CN122179329APending Publication Date: 2026-06-09QIANSAN (BEIJING) TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
QIANSAN (BEIJING) TECH CO LTD
Filing Date
2026-02-11
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

In existing technologies, terminal devices have limited computing resources, making it difficult to predict user behavior quickly and accurately. Edge devices have limited computing power, resulting in low prediction accuracy. Cloud feedback has high latency, leading to low efficiency in processing user behavior data.

Method used

A three-layer collaborative approach involving cloud, edge, and device is adopted. The terminal device performs preliminary data processing and transmits it to the edge device. The edge device uses the optimized model for prediction. If the confidence level is insufficient, it collaborates with other edge devices for processing. The cloud optimizes the large language model based on device capabilities and network conditions to generate a derivative model suitable for edge devices.

Benefits of technology

It improves the accuracy and response speed of user behavior prediction, reduces the computational burden on terminal devices, and enhances the model response speed and accuracy of edge devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122179329A_ABST
    Figure CN122179329A_ABST
Patent Text Reader

Abstract

This application relates to a cloud-edge-device three-layer collaborative user behavior prediction method, system, electronic device, and storage medium, comprising: collecting and processing user behavior data, sending the processed behavior data to a first edge device; receiving user behavior prediction results from the first edge device; the first edge device processing the processed behavior data using a built-in first behavior prediction model to obtain a first preliminary behavior prediction result and a first confidence score; when the first confidence score is greater than or equal to a score threshold, the first preliminary behavior prediction result is fed back as the user behavior prediction result; the first behavior prediction model is a derived model obtained by optimizing a large language model in the cloud based on the user behavior prediction task and the device capability data and network status data of the first edge device. The above method can improve user behavior accuracy and increase user response speed.
Need to check novelty before this filing date? Find Prior Art