数据处理方法、装置、设备及介质
By constructing feature vectors and relationship graphs for sample groups, and utilizing LSTM and graph attention neural network models, the accuracy problem of missing data completion in medical and health data was solved, achieving higher accuracy in data completion.
CN116994690BActive Publication Date: 2026-07-17PING AN TECH (SHENZHEN) CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PING AN TECH (SHENZHEN) CO LTD
- Filing Date
- 2023-07-06
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies have low accuracy when completing missing data in healthcare data, especially since traditional statistical and machine learning methods cannot effectively capture non-linear dependencies in the data.
Method used
By acquiring sample data, constructing sample groups, determining feature vectors and relationship graphs, and using pre-trained LSTM and graph attention neural network models, we can determine target feature vectors and perform data completion processing.
Benefits of technology
It improves the accuracy of data completion processing, can more accurately reflect the relationship between data, and improves the accuracy of missing data completion.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116994690B_ABST
Abstract
本发明涉及人工智能技术及医疗健康技术领域，公开了一种数据处理方法、装置、设备及介质，包括：获取第一样本数据；确定第一样本组；根据所述第一样本组，确定所述第一时序变量的第一特征向量，以及根据所述第一样本组，确定M个所述第二时序变量分别对应的第二特征向量；根据所述第一特征向量和M个所述第二时序变量分别对应的第二特征向量，确定第一关系图谱；根据所述第一关系图谱、所述第一特征向量和M个所述第二时序变量分别对应的第二特征向量，确定所述第一时序变量对应的第一目标特征向量；根据所述第一目标特征向量和所述第一特征向量，确定所述第一时序变量对应的第二数据。以提升对存在缺失数据的诊疗数据进行补全时的准确性。
Need to check novelty before this filing date? Find Prior Art