Business analytics (BA) is a term that is used to describe various approaches to data driven analysis including reporting tools such as OLAP and visualisation tools such as dashboards. For example the SAS Analytics software (www.sas.com/technologies/analytics) covers the following areas:]
- Statistics – use statistical data analysis to drive fact-based decisions.
- Data and text mining – build descriptive and predictive models and deploy results throughout the enterprise.
- Data visualisation – allows users to interact with graphs to clarify results and take action.
- Content categorisation – categorises content, which is then used to create metadata and trigger business processes
- Forecasting and econometrics – analyse and predict outcomes based on historical patterns and apply statistical methods to economic data, problems and trends
- Operations research – applies techniques such as optimisation, scheduling and simulation to achieve the best result
- Model management and deployment – streamline the process of creating, managing and deploying analytical models
- Quality improvement – identifies, monitors and measures quality processes over time.
Thus business analytical tools cover a wide range of techniques that includes data mining, text mining and web mining which are discussed earlier in this chapter. Various BA reporting tools such as OLAP and cube analysis, and BA visualisation tools such as dashboards and scorecards.
Online analytical processing (OLAP)
Online analytical processing (OLAP) refers to the ability to analyse in real time the type of large data sets stored in data warehouses. ‘Online’ indicates that users can formulate their own queries, compared to standard paper reports. The originator of OLAP, Dr E. Codd, defines it as the dynamic synthesis, analysis and consolidation of large volumes of multidimensional data. An example of a popular OLAP software application is shown in Figure 4.5. OLAP should not be confused with OLTP (online transaction processing). OLTP systems process large quantities of repetitive transactions conducting simple manipulations, whilst OLAP examines many data items in complex relationships. OLAP can be implemented on a relational database structure when it is termed ROLAP (relational OLAP) and the query
Online analytical processing (OLAP)
Online analytical processing (OLAP) refers to the ability to analyse in real time the type of large data sets stored in data warehouses. ‘Online’ indicates that users can formulate their own queries, compared to standard paper reports. The originator of OLAP, Dr E. Codd, defines it as the dynamic synthesis, analysis and consolidation of large volumes of multidimensional data. An example of a popular OLAP software application is shown in Figure 4.5. OLAP should not be confused with OLTP (online transaction processing). OLTP systems process large quantities of repetitive transactions conducting simple manipulations, whilst OLAP examines many data items in complex relationships. OLAP can be implemented on a relational database structure when it is termed ROLAP (relational OLAP) and the query
language SQL (see earlier in this chapter) can be used to interrogate the data. Alternatively OLAP can be implemented using a multidimensional database when it is termed MOLAP (multidimensional OLAP). OLAP queries can produce both routine reports that are generated automatically and periodically, for example weekly sales figures, and ad-hoc reports which are created in response to some event. An event can be triggered by a fall in sales volume or by a user request for specific information.
No comments:
Post a Comment