A data model for enhanced data comparability across multiple organizations

Abstract

Organizations may be related in terms of similar operational procedures, management, and supervisory agencies coordinating their operations. Supervisory agencies may be governmental or non-governmental but, in all cases, they perform oversight functions over the activities of the organizations under their control. Multiple organizations that are related in terms of oversight functions by their supervisory agencies, may differ significantly in terms of their geographical locations, aims, and objectives. To harmonize these differences such that comparative analysis will be meaningful, data about the operations of multiple organizations under one control or management can be cultivated, using a uniform format. In this format, data is easily harvested and the ease with which it is used for cross-population analysis, referred to as data comparability is enhanced. The current practice, whereby organizations under one control maintain their data in independent databases, specific to an enterprise application, greatly reduces data comparability and makes cross-population analysis a herculean task. In this paper, the collocation data model is formulated as consisting of big data technologies beyond data mining techniques and used to reduce the heterogeneity inherent in databases maintained independently across multiple organizations. The collocation data model is thus presented as capable of enhancing data comparability across multiple organizations. The model was used to cultivate the assessment scores of students in some schools for some period and used to rank the schools. The model permits data comparability across several geographical scales among which are: national, regional and global scales, where harvested data form the basis for generating analytics for insights, hindsight, and foresight about organizational problems and strategies.

Volume-Adaptive Big Data Model for Relational Databases

“…On the other hand, the software perspective is also promoting big data technologies such as single-sign applications with big data backends [5]. A big data model supporting a single-sign-on application that enhances data comparability across multiple organisations was designed and implemented by [16] To put volume in a perspective that emphasises its relevance to big data, volume is redefined by voluminosity, vacuum and vitality, three additional V-dimensions of data. These 3V's define the characteristics of volume in big data and are explained thus: i. Voluminosity -Voluminosity in volume states that there is a very large set of data collected so far and even much more is available to be collected.…”