Modeling and querying bitemporal semistructured data warehouses

Data Warehouse is very common and achieved prominence in recent years is a system that supplies more effective and rapid reports. Warehouse data is stored in a non-transactional repository. Easy reporting and reasoning via prepared warehouse cubes is ensured. It is important to understand customer?s tendencies and increase the customer satisfaction in corporations. At this point warehouse is a great solution. Temporal approaches are very important while analyzing historical data. Knowledge of old times is required on data analysis. To gain time- based realistic reports, changed data has to be stored in a certain order. Bitemporal approach is one of the temporal data storage approaches. Various methods may be used to store data in less space. Generating nested database tables is one of these methods. Nested tables compose a semi-structured format in database. The aim of this study is designing warehouse cubes which stores temporal values in bitemporal form, semi-structured format. Additionally cubes are queried and results are discussed. Outcomes may be used for analysis and data strategy development of companies. For supplying semi-structured form nested tables and nested types are used. 66 percent less disk space usage is provided. A dimensional database is designed. Queries are run in this dimensional database and results are discussed. Results are explained and supported with diagrams and charts. Keywords: Data warehouse, bitemporal data model, semi-structured data, Oracle warehouse builder.