Data processing method and device, electronic equipment, computer readable storage medium and computer program product
By clustering and determining the benchmark vectors for multiple first objects, the second sample data of the second object is enhanced by using the trained deep learning model, which solves the problem of low sample data diversity, and improves the efficiency of sample enhancement and the training effect of the model.
CN119719778APending Publication Date: 2025-03-28MASHANG CONSUMER FINANCE CO LTD
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411825231.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-11
- Publication Date
- 2025-03-28
AI Technical Summary
Technical Problem
In practical applications, the sample data used to train deep learning models is low in diversity, resulting in poor model training results.
Method used
By training the first model based on the first sample data of the plurality of first objects, a plurality of first clusters are obtained, and the reference vector of the second cluster is determined from it, the second sample data of the second object is enhanced by using the trained first model to generate the third sample data.
Benefits of technology
The efficiency of sample enhancement is improved, and the enhanced sample data is consistent with the original data in distribution, enriching the diversity of sample data and improving the training effect of the model.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention provides a data processing method and device, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: training a first model based on first sample data of a plurality of first objects to obtain a trained first model; clustering the plurality of first objects based on the first sample data to obtain a plurality of first clusters; a second object is determined from the multiple first objects, a second cluster is determined from the multiple first clusters, the second object is any first object, and the second cluster is the first cluster where the second object is located; determining a reference vector of the second cluster based on the first sample data corresponding to the first object contained in the second cluster; and based on the reference vector, performing enhancement processing on second sample data of the second object through the trained first model to obtain third sample data. According to the invention, the sample enhancement efficiency can be improved, and the diversity of the sample data can be improved.
Need to check novelty before this filing date? Find Prior Art