A distributed big data fusion system and method for multi-source heterogeneous data
By using a distributed big data fusion system to collect data pattern metadata and map model alignment from multi-source heterogeneous data, multi-dimensional feature extraction and similarity matrix calculation are achieved. This generates associated data clusters and performs distributed computation, solving the computational bottleneck and consistency issues in multi-source heterogeneous data processing. It also improves the flexibility and accuracy of data fusion and ensures the precision and transparency of globally fused data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SICHUAN VOCATIONAL COLLEGE OF CHEM TECH
- Filing Date
- 2026-05-21
- Publication Date
- 2026-07-10
AI Technical Summary
Traditional methods struggle to handle multi-source heterogeneous data due to issues such as data mismatch, information loss, semantic inconsistency, computational bottlenecks, low processing efficiency, and poor scalability. Furthermore, they lack tracking records of data sources and processing procedures.
A distributed big data fusion system is adopted to collect data pattern metadata and align the mapping model, perform multi-dimensional feature extraction and similarity matrix calculation, generate associated data clusters, and perform distributed computing node allocation and global weighted aggregation to achieve data standardization and consistent fusion.
It improves the flexibility of data access and fusion, quantifies the similarity between data entities, enhances the accuracy and completeness of data association, ensures the accuracy and stability of globally fused data, and improves the transparency and manageability of the data fusion process.
Smart Images

Figure CN122365393A_ABST