Utilization Extract, Transform, Load For Developing Data Warehouse In Education Using Pentaho Data Integration

The utilization of data warehouses in various fields is an absolute necessity. A data warehouse is a database that contains large amounts of data that aims to help organizations, fields, and institutions specifically for decision making. Data warehouses can produce important information in the future. Loading data from various sources and processed through an ETL (Extract, Transform, Load) process that displays data consistently is the basis for creating a data warehouse architecture. The development of a data warehouse in education will provide significant benefits for the progress of education. Integration of data and processing results stored in the data warehouse can be the basis for evaluating better planning. Development of data warehouse adopt the multidimensional modelling method which consists of four stages: select the business process, declare the grain, select dimensions, and identify facts. This stage produces a data warehouse architecture and influences and contributes to the advanced information technology in education.

UML-Based Data Warehouse Design Using Temporal Dimensional Modelling

The design of a data warehouse system deals with tasks such as data source administration, ETL processing, multidimensional modelling, data mart specification, and end-user tool development. In the last decade, numerous techniques have been presented to cover all the aspects of DW. However, none of these techniques stated the recent necessities of DW like visualization, temporal dimensions, record keeping, and so on. To overcome these issues, this paper proposes a UML based DW with temporal dimensions. This framework designs time-dependent DW that allows end-users to store history of variations for long term. Besides, it authorizes to visualize the business goals of organizations in the form of attribute tree via UML, which is designed after receiving user necessities and later reconciling with temporal variables. The implementation of proposed technique is detailed with university education database for quality improvement. The proposed technique is found to be useful in terms of temporal dimension, long-term record keeping, and easy to make decision goals through attribute trees.

Hilbert space multidimensional modelling of continuous measurements

Data fusion problems arise when a researcher needs to analyse results obtained by measuring empirical variables under different measurement contexts. A context is defined by a subset of variables taken from a complete set of variables under investigation. Multiple contexts can be formed from different subsets, which produce a separate distribution of measurements associated with each context. A context effect occurs when the distributions produced by the different contexts cannot be reproduced by marginalizing over a complete joint distribution formed by all the variables. We propose a Hilbert space multidimensional theory that uses a state vector and measurement operators to account for multiple distributions produced by different contexts. This article is part of the theme issue ‘Contextuality and probability in quantum mechanics and beyond’.