Article Preview
TopIntroduction
E-business domains, such as E-commerce, E-government, E-learning, E-banking, are growing fast, and with this growth, companies are willing to spend more money and time to analyze the data behind their websites. Based on our literature review, E-business companies have been using either primitive measures to analyze their data or more sophisticated analysis techniques, such as DM or OnLine Analysis Processing (OLAP), to generate useful data and knowledge for decision makers (Kohavi & Provost, 2001; Kohavi et al., 2004). DM and OLAP techniques have shown interesting results, but they present some restrictions concerning the size, the structure, and the type of the data to be analyzed, as well as the level of expertise of end-users who will be interpreting the results. With growing pressure to make E-business companies more profitable, additional analysis techniques are required to analyze the data which are becoming more and more complex.
Current studies have examined the effect of what is called Business Intelligence (BI), which is a combination of data analysis techniques such as DM with OLAP, to analyze fused data coming from different sources. Some researchers have used only textual and numerical data (Codd et al., 1993; Youngworth, 1995; Lau et al., 2008; Zhou et al., 2009) and ignored the geographical features of the data which are presented through text and numbers. In these studies, the analysis lacks visualization or mapping of the analysis results. Other researchers have conducted data analysis based on spatial (geographical) data only, which results in limited and incomplete data analysis (Kouba et al., 2000; Stefanovic et al., 2000; Ferreira et al., 2001; Shekhar et al., 2001; Fidalgo et al., 2004; Bédard et al., 2005; Scotch & Parmanto, 2005; Silva et al., 2005).
Although one cannot deny the importance of previous studies, limited research has considered integrating multiple data analysis techniques to improve the process of decision making and provide comprehensive, easier and friendlier data analysis in an E-business domain. Decision makers, in this rapidly changing world, are demanding faster and more detailed results. To address this issue, more research is needed in investigating the infusion of GIS (Geographic Information System), spatial analysis and non-spatial analysis applications and tools in business. Thus, our research addresses the following question: how can applying different useful analysis techniques to analyze fused and complex data coming from different sources and using different formats (spatial and non-spatial) benefit decision makers? In other words, will the integration of multiple data analysis techniques be more useful and more appropriate in the decision-making process?
In our research, we examine how the infusion of different analysis techniques may be applied in order to help decision makers make better decisions. Thus, we suggest the integration of the following well-known data analysis techniques: OLAP and SOLAP (Spatial OLAP) (for spatial data) in order to analyze fused data coming from E-business websites (non-spatial data) and GIS (spatial/geographical data). First, we have developed a framework which integrates the OLAP and SOLAP analytical techniques to analyze fused data coming from different sources (E-business databases and GIS) to perform comprehensive, easier, and more advanced and sophisticated analyses of E-business websites’ databases. Then, we have tested this framework using actual data from an E-business website related to online job seekers in the UAE. Our study provides more comprehensive data analysis of numerical, textual, and spatial data, which can help the decision makers to make better decisions. Based on our review of literature, no similar framework has been proposed, which makes our framework unique and can be considered as a contribution to the fields of E-business, data analysis, and BI.