This work is supported by National Key Research and Development Program of China (2019YFB2102500)
政务数据资源来源广泛、类型多样、数据量大且分布情况不清晰，缺乏统一管理，数据获取和使用效率较低，无法持续释放数据价值。为解决上述问题，实现不同来源和不同类型数据的关联融合，该文采用多源异构数据融合技术，为政务数据的交换和共享提供技术支持。该文不仅提出了信息通道和通道耦合的概念，还提出一种基于知识图谱的多源异构信息通道耦合的方法，可先实现多源异构数据统一化，再进行数据融合。该方法包括构建初始通道耦合知识图谱、基于通道耦合知识图谱实现通道数据耦合和基于通道耦合知识图谱实现知识更新 3 个模块。基于知识图谱的多源异构信息通道耦合的方法将图谱构建、知识抽取、知识融合、知识加工、知识更新等技术相结合，通过一种自顶向下的排序检索模型，加快了知识融合和数据检索的速度和准确度。
Government data resources are characterized by a wide range of sources, diverse types, large data volumes and unclear data distribution, which can lead to lacks of unified management, high efficiency in data access and use, and ability to continuously promote the data release value. In order to solve the above problems and realize the correlative fusion of data from different sources and different types, this paper adopts the fusion of multi-source heterogeneous data to provide technical means for the exchange and sharing of government data. This paper proposes the concept of information channel and channel coupling, and provides a multisource heterogeneous information channel coupling method, which includes three module for constructing the Initial channel coupling knowledge graph, realizing the channel data coupling based on the channel coupling knowledge graph, and realizing the knowledge updating based on the channel coupling knowledge graph, respectively. In the proposed method, the technologies of atlas construction, knowledge extraction, knowledge fusion, knowledge processing and knowledge updating are integrated together, and a top-down sorting retrieval model is proposed to accelerate the speed and accuracy of knowledge fusion and data retrieval.
HAN Chengshan, LI Honghui, YAN Jiahe, et al. Research on Coupling Technology of Multi-source Heterogeneous Information Channels Based on Knowledge Graph[J]. Journal of Integration Technology,2023,12(3):48-60