Data visualization method, device, system, equipment and medium using AI model

By extracting and fusing features from multi-source heterogeneous auto insurance data, and utilizing auto insurance attention mechanisms and high-dimensional semantic AI annotation models, the problem of data feature loss in auto insurance visualization has been solved, thereby improving data accuracy and display effects.

CN121837417BActive Publication Date: 2026-06-09SHENZHEN TUOBAO SOFTWARE CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHENZHEN TUOBAO SOFTWARE CO LTD
Filing Date
2026-03-13
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies process data of different dimensions in the car insurance scenario through dimensionality reduction, which leads to the loss of the core features of the data. This results in large deviations between the visualized data and the actual business scenario, low accuracy, and poor display effect.

Method used

We acquire multi-source heterogeneous auto insurance company data, including structured policy data, unstructured text and image data, and sensor data. After feature extraction and standardization, we construct auto insurance visualization charts through an auto insurance attention mechanism fusion model and a high-dimensional semantic AI annotation model.

Benefits of technology

Ensure data integrity, improve the accuracy and effectiveness of visualized data, make it fit the actual business scenarios of auto insurance, and avoid feature loss caused by traditional dimensionality reduction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121837417B_ABST
    Figure CN121837417B_ABST
Patent Text Reader

Abstract

The application discloses a data visualization method, device, system, equipment and medium using an AI model, and belongs to the field of data visualization. The application can acquire vehicle insurance policy data, vehicle insurance text and image data, and sensor data. Non-structural feature vectors are extracted from the vehicle insurance text and image data, the vehicle insurance policy data is converted into structural feature vectors, and alignment data is extracted from the sensor data. A fusion weight value is determined based on the numerical size of the alignment data, a vehicle insurance attention mechanism fusion model is called, the non-structural feature vectors and the structural feature vectors are fused according to the fusion weight value, and vehicle insurance fusion feature vectors are obtained. An AI labeling model is called to label vehicle insurance exclusive business labels on the vehicle insurance fusion feature vectors, vehicle insurance visualization charts are constructed based on the vehicle insurance exclusive business labels, and the vehicle insurance visualization charts are subjected to visualization processing. The application can avoid feature loss caused by traditional dimension reduction, so as to improve the accuracy and effect of the visualization data.
Need to check novelty before this filing date? Find Prior Art