Study On Heterogeneous Data Information Integration Of Shenhua Ningxia Coal Group

Shenhua Ningxia Coal Group is one of the top coal industries leading the technology in information construction.At present,it has 23 sets of management information systems.These systems are playing an important role in the whole group.However,these systems are independent with each other,and they are totally different in the development platforms,operating systems and databases.With the constant development of informatization construction,it becomes eager for the group to unify the scattered databases,and further linking those independent systems.In order to achieve data sharing,it is highly urgent to make full use of the dispersed data in different systems.In this case,the Shenhua Ningxia CoalGroup proposed to make a research on heterogeneous data integration.Firstly,this paper introduces the classical technology of data integration and further make a discussion about their details.We all know that Extensive Markup Language(XML)is a type of cross-platform language.so data formats can be exchanged between different databases through XML.In another word,it is possible to solve the problem of independent heterogeneous databases by using the middleware methods based on XML.The advantage of adopting middleware is that it can shield the heterogeneity between databases and share the data between systems.And then,we can achieve the goal to fully use data and share data easily.Secondly,after analyzing the present status of information construction in Shenhua Group,we proposed a kind of data integration model,the model is based on the database middleware.What's more,this model can achieve data sharing without changing the original ways in storing and managing the information of heterogeneous databases.We deal with the data from different systems by adding middleware between the layer of database and the layer of presentation.The middleware layer is added through the application layer and the data layer to break the heterogeneity between databases so as to achieve data sharing between systems.Then,according to the problem of data heterogeneity and semantic heterogeneity,the steps of integration between databases are given.In the data integration process,in order to solve the semantic heterogeneity among heterogeneous data sources,an attribute column matching algorithm between database tables is proposed.First,the attribute columns are classified by type and the classified result sets are calculated by similarity matching method.The similarity value between the attribute columns is matched with a value with a high degree of similarity.Finally,the proposed method is applied to the "Realization of ERP-SAP Information Integration for Shenhua Ningxia Coal".The system has been officially running in the enterprise and is running smoothly.Practice has proved that the method proposed in this paper is effective and feasible.