Abstract:
In data integration, users are provided with a homogeneous logical view of data that is physically distributed over heterogeneous data sources. However, there exists elements of ”Chaotic Information Systems” in most institutions as a result of the presence of ad-hoc incompatible information systems comprising of fragmented mixtures of single function systems that were built to meet certain operational requirements of the user departments and therefore neither share data nor communicate effectively with other sub-systems within the organization. But if a unified data warehouse exists in an academic institution, then that constraint of viewing only the data a unit has will be solved. A typical tertiary institution often comprises a lot of subsystems crucial for its internal processes and operations. Therefore this study presents a data integration model for the management information system in a tertiary institution. Auchi Polytechnic is used as a case study to test the viability of the model with the integration of data from 3 units (Admission, Result Processing and Bursary) called local sources. The Local-As-View (LAV) data integration approach is used, and treating all local sources as relational databases. Schematic mappings were used to map and create relationship between the global (mediated) schema and the local sources in order to form the unified view. A mediator in the proposed system acts as an intermediate medium transforming queries issued by users/clients to sub-queries, integrating result data and resolving conflicts and inconsistencies. Wrappers transform sub-queries to specific local queries so that each local system is able to understand the queries to get expected results. The model is implemented using web-enabling applications with apache as a web server with extended capability for PHP scripting language and MySQL database. The system was tested bythree sets of users of the system: Frequent users (6) who will use the system on daily basis, Occasional users (7) who use the system say monthly and Rare users (4) who may use the system yearly. The system was evaluated using a questionnaire that was filled by these users (IT and non-IT staff of Auchi Polytechnic, Auchi). From their responses that were analysed along side the results/output from the system, it was concluded that the data integration system was viable as the outputs obtained from the issued queries to the system were precise due to the fact that the result of the precision metrics shows that mean is 71.5% and Range is 71.0%. This model can be adapted by any tertiary institution, only just to con it to their operational system.