Requirements analysis method for extracting-transformation-loading (ETL) in data warehouse systems

• Main Authors: Ta'a, Azman, Abdullah, Mohd Syazwan, Md Norwawi, Norita
• Format: Conference or Workshop Item
• Language: English
• Published: 2009
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/13500/1/PID97.pdf; http://repo.uum.edu.my/13500/; http://www.icoci.cms.net.my

The data warehouse (DW) system design involves several tasks such as defining the DW schemas and the ETL processes specifications, and these have been extensively studied and practiced for many years.The problems in heterogeneous data integration are still far from being resolved due to the complexity of ETL processes and the fundamental problems of data conflicts in information sharing environments.The understanding of an early phase of DW development is essential in properly tackling the complexity of ETL processes.The method to analyses the DW requirements from the abstract level (e.g. goal, sub-goal, stakeholder, dependency) toward the specification of ETL processes (e.g. extracting, filtering, conversion) are important in order to manage the complexity of the ETL processes design (e.g. semantic heterogeneity problems). However, current approaches that are based on existing software requirement approach still have limitations on translating the business semantics for DW requirements toward the ETL processes specifications.Moreover, the understanding of goal in the perspective of the organization and decision makers are important to ensure the semantic of DW requirements can be properly determined, organized, and implemented by the ETL processes. Therefore, the proposed method will utilize the ontology with goal-driven approach in analyzing the requirements of ETL processes.