Multi-heterogeneous data fusion for enterprise data asset valuation in public health policy context.

Introduction: In the dynamic realm of public health policy, the valuation of enterprise data assets is increasingly pivotal. This paper introduces a comprehensive methodological framework for multi heterogeneous data fusion, aimed at enhancing data asset valuation in this context. Traditional methods often struggle with integrating diverse data types, structured, unstructured, and semi structured, into a cohesive analytical structure. Methods: To address this challenge, this study proposes a novel framework composed of FusionNet and an Innovative Fusion Strategy. FusionNet employs machine learning techniques to fuse varied data sources, while the Innovative Fusion Strategy aligns data integration with policy oriented objectives. The framework further incorporates an Adaptive Data Synthesis mechanism that applies domain knowledge to optimize valuation accuracy. Results and discussion: Experimental results on four public health datasets demonstrate that the proposed method improves valuation accuracy by 3.6% and AUC by 2.2% compared to strong baselines. These findings suggest the proposed framework enhances decision support for public health policy through improved data integration and valuation modeling.