PhD Defence: Explanation of Exceptional Values in Multi-dimensional Business Databases


Explanation of Exceptional Values in Multi-dimensional Business Databases

Every day, business analysts create and manage piles and piles of data from all kinds of business operations. Yet even such huge amounts of data don’t give the right information for making management decisions. In his dissertation entitled 'Explanation of Exceptional Values in Multi-dimensional Business Databases', PhD candidate Emiel Caron has revealed an innovative method to integrate explanatory business analytics into existing decision support systems. His findings not only expose exceptional values in business data, but also explain their underlying causes.

Information managers often have the contents of and statistics from transactions, but don’t have the answers to simple questions such as why sales were better in March than in February. Multi-dimensional or OLAP (OnLine Analytical Processing) databases already enable users to easily and selectively extract and view data from different points of view. However, Caron’s technique not only supports the discovery of exceptional values in OLAP data but adds the explanation of such values by providing their underlying causes. Potential uses include order entry applications, customer relationship management (CRM) applications and warehouse management systems.

In his PhD dissertation Explanation of Exceptional Values in Multi-dimensional Business Databases, Caron describes a newly found method that can automatically detect exceptional values at any level in the data using statistical models. Secondly, the programme forms a generic model for diagnosis of atypical values, by showing a full explanation tree of causes at successive levels.

If the tree is too large, the data analyst can use appropriate filtering measures to ‘prune’ the tree to a manageable size. Caron’s methodology also offers a wide range of applications such as inter-firm comparison, analysis of sales data and the analysis of any other data that possess a multi-dimensional hierarchical structure. The use of Caron’s analytical system could offer companies a better insight into exceptional values, quickly provide the assignable causes – and thus advance business.

Caron defended his dissertation  on Thursday 14 November 2013. His supervisor was Professor Hennie Daniels and his co-supervisor was Professor George Hendrikse. Other members of the Doctoral Committee were Professor Bert Balk (Erasmus University Rotterdam), Professor Uzay Kaymak (Technische Universiteit Eindhoven) and Professor Jos van Hillegersberg (University Twente).

About Emiel Caron

Emiel Caron (the Netherlands, 1975) obtained his master's degree in Information Management, with a specialisation in Information Technology (IT), from Tilburg University in 2000. In the same year he started to work as a business intelligence (BI) consultant for PinkRoccade (now KPN) on data warehousing and BI projects. In 2002, he started his PhD research at the Decision & Information Sciences department of RSM. Later he joined the former department of Economics & Informatics at the

Erasmus School of Economics as an assistant professor and lecturer in business information systems. Caron's PhD research is in the field of business intelligence and analytics and focuses on the explanation of exceptional values in multi-dimensional databases.

In his PhD trajectory he received two research grants from the Netherlands Organisation for Scientific Research to work with the Natural Computing Research & Applications Group at University College Dublin. Results of his research have been presented at various international conferences and have been published in various conference proceedings and journals (European Journal of Operational Research, Intelligent Systems in Accounting, Finance and Management, and International Journal of Business Intelligence Research). He is also a member of the Erasmus Centre for Business Intelligence.

Since 2012, Caron has worked as an IT developer and researcher at the Centre for Science and Technology Studies of Leiden University. He has become interested in the application of business intelligence and data mining techniques on large scientometric and bibliometric databases. Moreover, he teaches the course business intelligence in the master ICT in Business of the Leiden Institute of Advanced Computer Science and supervises several MSc students.

Abstract of Explanation of Exceptional Values in Multi-dimensional Business Databases

Multi-dimensional or OnLine Analytical Processing (OLAP) databases are a popular business intelligence technique in the field of enterprise information systems for business analytics and decision support. In his dissertation, Caron extends OLAP database functionality to support the business analyst in the exploration of OLAP data. The database is augmented with novel functionality for the detection of exceptional values, explanation generation, and sensitivity analysis. Caron describes how exceptional values at any level in the data, can be automatically detected by statistical and managerial models. Also shown is how exceptional values can be explained by underlying causes. This is realised by a generic model for diagnosis of atypical values. By applying it, a full explanation tree of causes at successive levels can be generated. If the tree is too large, the analyst can use appropriate filtering measures to prune the tree to a manageable size. The purpose of the methods and algorithms presented is to provide OLAP databases with more powerful explanatory analytics and reporting functions. This methodology has a wide range of applications, such as variance analysis in accounting, competition benchmarking, analysis of sales and financial data, and the analysis of any other data that possess a multi-dimensional hierarchical structure. The method is demonstrated in several case studies. For example, the explanatory analysis of a sales data cube is discussed, and computerised competition benchmarking with financial data about Dutch retail companies is illustrated.

Photos: Chris Gorzeman / Capital Images