| Summary: | This study aims to achieve intelligent fusion and unified modeling to meet the requirements of multi-source and heterogeneous big data granulation for knowledge discovery in the field of water resources. The paper focuses on decision-making data granulation and knowledge discovery driven by big data in the field of water resources. It utilizes a combination of domain numerical simulation and model verification to systematically investigate decision-oriented big data multi-granularity granulation and knowledge discovery. The study reveals the mechanism and law of the transformation of management and decision-making paradigm driven by big data. This study results include the development of a granulation mechanism and a semantic fusion method for multi-source and heterogeneous big data, a multi-scale granular structure for big data, multi-granularity feature discovery and granulation method, and a multi-granularity uncertainty reasoning and knowledge discovery method. The proposed dynamic big data fusion and knowledge discovery approach effectively supports big data granulation and knowledge discovery in water resource decision-making. The study found that the proposed dynamic big data multi-granularity fusion method outperforms existing dynamic big data correlation analysis methods and greatly reduces data processing time.
|
|---|