Data processing method based on multi-source data fusion and dynamic threshold correction

By using multi-source data fusion and dynamic threshold correction, the target object can be quickly and accurately screened from multi-source heterogeneous data, solving the problem of low data object screening efficiency in existing technologies and improving the automation and accuracy of data processing.

CN122432222APending Publication Date: 2026-07-21SHUOHUANG RAILWAY DEV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHUOHUANG RAILWAY DEV
Filing Date
2026-04-10
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

In existing technologies, data object screening is inefficient, relies on human experience leading to long processing times, and cannot quickly filter out target objects that meet the criteria from a large amount of data.

Method used

By using multi-source data fusion and dynamic threshold correction, multi-source heterogeneous data is collected and subjected to cross-dimensional normalization processing. A suitable prediction model is selected for scoring prediction. The dynamic screening threshold is determined by combining the current batch and historical benchmark data, thereby realizing automatic screening of data objects.

Benefits of technology

It improves the efficiency and accuracy of data object screening, avoids the evaluation bias of a single model, ensures that the screening criteria are adapted to the actual data distribution, and enhances the automation and accuracy of data processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122432222A_ABST
    Figure CN122432222A_ABST
Patent Text Reader

Abstract

The application relates to a data processing method and device based on multi-source data fusion and dynamic threshold correction, computer equipment, a computer readable storage medium and a computer program product, which can be used in the technical field of computers. The method comprises the following steps: performing cross-dimension normalization processing on multi-source heterogeneous data to obtain standardized feature data of a target data object; selecting a corresponding target prediction model from a plurality of differentiated prediction models; inputting the standardized feature data into the target prediction model to perform scoring prediction processing and obtain a prediction score; acquiring distribution characteristics of the prediction scores of all data objects in a current processing batch; determining a dynamic screening threshold corresponding to the current processing batch according to the distribution characteristics and pre-stored historical reference data; and determining a data object screening result corresponding to the target data object according to a screening condition corresponding to the prediction score of the target data object and the dynamic screening threshold. The method can improve the efficiency of data object screening.
Need to check novelty before this filing date? Find Prior Art