A Research On Domain-related Increamental Data Integration Method

With the advancement of business intelligence technology,the rapid development of informatization has brought about the problem of “information isolated island” where information systems are diversified and data cannot be shared.Therefore,it is urgently necessary to proceed from an overall perspective of planning,to integrate the originally scattered and diverse information resources from the physical or logical data integration in information systems,to break the “information isolated islands” and realize the sharing of information resources.First of all,for the issues of that there are many information systems and data can not be shared in enterprises and government units,a domain-related incremental data integration scheme is proposed.The domain-related incremental data integration framework includes three main parts,namely domain information resource integration,incremental data extraction module and data set fusion module.The integration of domain information resources mainly involves the analysis of the services of each functional department and the standardized description of the functional office resources from the functional domain,business process,and data elements.Secondly,an incremental data extraction module is proposed to solve the problem of the existing integration technology for large-scale and high-cost reconstruction of the established information system.The module is based on the network sniffer technology and solves the problem of real-time access to the dynamic information resources of the service office without affecting the operation of the existing information system.It operates in the form of local services and operates according to the user's database operations.Returning the query result set from the database server reduces the cost of manual intervention.Thirdly,for the problem of large number of data sets,multiple data types,and low value density,a multi-dataset fusion model was proposed.First,a single data set is preprocessed,and a priori description and an example description are described.Then the correlation relationship between the data sets is analyzed,and the extracted single data sets are fused into one data set.Finally,using the hotel occupancy rate forecast as the target decision-making demand,using the incremental data extraction method and data collection to collect the provincial tourism commission's partial office data set to fuse,and verify the validity of the method.