A Value-based approach to modelling interoperability in Knowledge Organization Systems

Abstract

This study will identify the mechanisms that are required within the context of enterprise application to achieve a value-based level of system interoperability. This will be translated to provide useful insight on how to optimize interoperability in Knowledge Organization Systems (KOSs) using Information System Modelling Approach (ISMA). To do this, the design and implementation of a unified value-based model and its evaluation will be proposed based on the mechanisms identified. The work will focus on the integration of enterprise data as information at the level of conceptual modelling following the provisions in Diniz et al. (2006) to benefit KOSs in terms of interoperability with the goal of achieving information access. Interestingly, existing interoperable KOS model is still evidence-based with focus on how well it has improved and holding up. On the contrary, a value-based approach will seek the maturation of an interoperable model that will significantly assists its users in meeting their information need to derive satisfaction. The goal is to leverage how enterprise systems efficiently incorporate third-party applications to provide its users with a unified access to information. From a software engineering perspective, a value-based interoperable model will be useful to optimize the complex integration of information that KOS needs to provide a “one-stop” seamless search process. The complexity of the integration is such that just by entering one query, a user should be able to retrieve results from various heterogeneous resources (Zeng et al., 2004). Basically, this means that the components of different KOSs - controlled vocabularies and classification schemes - are to be made interoperable. What is in contention therefore, is that within a particular search apparatus the exchange and use of information should be without special effort on the part of integrating systems. This paper will centre around the primacy of correctness and reusability of an integrative modelling of conformance in terms of requirements and database re-engineering, and metadata restructuring for interoperable purpose. The mechanisms as components required for the value-based model design will be identified using appropriate fact-finding techniques like active stakeholder’s participation, interviews, record inspection, direct observation and search from secondary sources. The implementation will be done with re-engineering technique like Extract-Transform-Load (ETL) mechanism. Based on potential usage, the various modules of integrating a third-party value-based model will be specified by categorising all extracted information to showcase seamless information access. The proposed model design will be done using the Unified Modelling Language tools. The rational unified process methodology will be introduced to incorporate the dimension of software evolution process to satisfy the requirement of interoperability. It will be needful to integrate the stages of business process modelling - ISMA, and scrutinize the proposed design to test it. With this template, an object-based and Web-empowered methodology that uses the Microsoft Structured Query Language (SQL) Studio, Active Server Pages (ASP.net) Web services, and Procedural Language/SQL codes will be employed to provide the needed maturation in this work. The foregoing will allow the re-engineering of tightly coupled enterprise database structure. The customer relationship management system - a type of KOS – will serve as a case study. Its dynamics will be drawn on to achieve interoperability using the SQL Server Integration Services algorithmic technique to manage all SQL server-side integration services. At the level of information access this will practically ensure inductive inferences after an interoperable session over supervised information as data. The value-based model will be evaluated based on the white box technique to understand and highlight KOSs functional requirements from the perspective of orientating value addition. This should ensure support for interferences and adoptions that encourage interoperability. Historical data from different flat files will be used to test the proposed model’s ability to analyse grouped data that are aggregated into different formats with views created in hypercube. This will guarantee better outcome in the long term with value-based modelling approach when compared to that of evidence-based. The result of the foregoing will demonstrate the possibility of seamlessly re-engineering legacy applications like KOSs. For KOSs, the result that will be arrived at will show that a universal integration layer can foster interoperability that gives room for more success at the seamless integration of disparate information sources. It will also highlight the fact that since there is a reasonable amount of data processing activities when KOSs are integrated; from the angle of Webservices, interoperability can still be guaranteed for KOS since XML scripts can be re-engineered to intervene. Furthermore, the result will also show that huge volumes of data that are generated from various operational systems; if they exist can easily be accessed. Therefore, the study will conclude that with the creation of a universal layer as an interoperable platform, disparate sources of information will be easily accessed. The overarching consequence in this, is that with every system integration comes the implication of discovering unseen relationships that exist in data and text. And this can be re-oriented towards better interoperability for easy information access. Therefore, the value-based model will show that interoperability in KOSs cannot be compacted as just intervention to satisfy information access alone. However, it must include a wider perspective that considers outcomes such as utility, satisfaction, user preferences, and level of impact on users’ function haven had access to the right information. Keywords: Value-based modelling, Information system modelling, Interoperability, System integration, Enterprise-applications, Re-Engineering, Knowledge organization systems References Diniz, B., Nogueira, D. L., Cardoso, A., Ferreira, R. A., Guedes, D., and Meira, W. (2006). Assessing Data Virtualization for Irregularly Replicated Large Datasets. In Proceedings of the Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06) (Vol. 1, pp. 505-512). Zeng, Marcia Lei, & Chan, Lois Mai. (2004). Trends and Issues in Establishing Interoperability among Knowledge Organization Systems. J o urnal of th e A m eri c an Soc iet y f o r Inf o rmation Science and Te c hn o l o gy , 55(5): 377-395.