Association Rules Mining Regarding the Value of Business Intelligence Solutions
Association Rules Mining Regarding the Value of Business Intelligence Solutions
Author(s): Petr Havel, Manomeet Gupta, Athanasios PodarasSubject(s): ICT Information and Communications Technologies
Published by: UIKTEN - Association for Information Communication Technology Education and Science
Keywords: association rule mining; business intelligence importance; Apriori Algorithm; Rpackage
Summary/Abstract: The paper investigates the importance of business intelligence solutions in modern enterprises using association rule mining techniques. The research is based on a questionnaire addressed to different employee target groups regarding their age interval, their employment status, their domain of employment, their experience or inexperience with business intelligence tools and their positive or negative aspect regarding the importance of business intelligence in modern companies. 90 responses have been received and used for dataset formulation. Using the association rule induction standard procedure, the most popular rules with respect to different antecedent item combinations and business intelligence value as consequent item have been inferred setting as minimum confidence 50% and minimum support 0,1. The collected data have been prepared in common separated values format and the association rules have been inferred using the R- Package. In general, among other rules, a strong relation between BI experience and positive BI aspect can be reported which is also confirmed via simple Pearson X2 statistical test in R. An investigation paradox which has been spotted is the negative opinion regarding the BI usefulness stemming from a minority of respondents familiar with BI tools.
Journal: TEM Journal
- Issue Year: 11/2022
- Issue No: 3
- Page Range: 1399-1405
- Page Count: 7
- Language: English