Application of Data Mining techniques to identify relevant Key Performance Indicators

作者：Peral Jesus^*; Mate Alejandro; Marco Manuel

来源：Computer Standards & Interfaces, 2017, 54: 76-85.

DOI：10.1016/j.csi.2016.11.006

• KPIs
• Data Mining
• Big data
• Decision trees
• Open Data

Currently dashboards are the preferred tool across organizations to monitor business performance. Dashboards are often composed of different data visualization techniques, amongst which are Key Performance Indicators (KPIs) which play a crucial role in quickly providing accurate information by comparing current performance against a target required to fulfil business objectives. However, KPIs are not always well known and sometimes it is difficult to find an appropriate KPI to associate with each business objective. In addition, Data Mining techniques are often used when forecasting trends and visualizing data correlations. In this paper we present a new approach to combining these two aspects in order to drive Data Mining techniques to obtain specific KPIs for business objectives in a semi-automated way. The main benefit of our approach is that organizations do not need to rely on existing KPI lists or test KPIs over a cycle as they can analyze their behavior using existing data. In order to show the applicability of our approach, we apply our proposal to the fields of Massive Open Online Courses (MOOCs) and Open Data extracted from the University of Alicante in order to identify the KPIs.

• 出版日期2017-11