Capturing high-level requirements of information dashboards' components through meta-modeling

Abstract

Information dashboards are increasing their sophistication to match new necessities and adapt to the high quantities of generated data nowadays. These tools support visual analysis, knowledge generation, and thus, are crucial systems to assist decision-making processes. However, the design and development processes are complex, because several perspectives and components can be involved. Tailoring capabilities are focused on providing individualized dashboards without affecting the time-to-market through the decrease of the development processes' time. Among the methods used to configure these tools, the software product lines paradigm and model-driven development can be found. These paradigms benefit from the study of the target domain and the abstraction of features, obtaining highlevel models that can be instantiated into concrete models. This paper presents a dashboard meta-model that aims to be applicable to any dashboard. Through domain engineering, different features of these tools are identified and arranged into abstract structures and relationships to gain a better understanding of the domain. The goal of the meta-model is to obtain a framework for instantiating any dashboard to adapt them to different contexts and user profiles. One of the contexts in which dashboards are gaining relevance is Learning Analytics, as learning dashboards are powerful tools for assisting teachers and students in their learning activities. To illustrate the instantiation process of the presented meta-model, a small example within this relevant context (Learning Analytics) is also provided.

Representing Data Visualization Goals and Tasks through Meta-Modeling to Tailor Information Dashboards

“…The dashboard meta-model is also part of the four-layer meta-model architecture proposed by the OMG, in which a model at one layer is used to specify models in the layer below [39]. In particular, the first version of the dashboard meta-model [20] was an instance of MOF (i.e., an M2-model), so it can be instantiated to obtain M1-models. This meta-model was transformed in an instance of Ecore [40] using Graphical Modelling for Ecore included in Eclipse Modeling Framework (EMF), in order to leverage the different features of this modeling framework (Figure 1).…”

“…This paper discusses the main factors that need to be accounted for in dashboard design and extends a dashboard meta-model that identifies core relationships and entities within this complex domain [18][19][20]. The previously developed meta-model aimed at formalizing a structure for defining information dashboards based on a set of factors, such as the data structure or users' goals, preferences, domain knowledge, visualization literacy, etc.…”

A Meta-Model Integration for Supporting Knowledge Discovery in Specific Domains: A Case Study in Healthcare

“…The dashboard meta-model is also part of the four-layer meta-model architecture proposed by the OMG, in which a model in one layer is used to specify models in the layer below [52]. In particular, the first version of the dashboard meta-model [47,48] was an instance of MOF (i.e., an M2-model), so it can be instantiated to obtain M1-models. This meta-model was transformed in an instance of Ecore [36] using Graphical Modelling for Ecore included in the Eclipse Modeling Framework (EMF), in order to leverage the different features of this modeling framework (Figure 1).…”

“…As introduced in the first section, the holistic meta-model presented in this work merges two previously developed meta-models: a learning ecosystem meta-model [36] and a dashboard meta-model [47,48]. This dashboard meta-model has a finer grain than the aforementioned meta-models present in the literature, allowing more sophisticated combinations to obtain a wide variety of dashboard displays.…”

A Meta-modeling Approach to Take into Account Data Domain Characteristics and Relationships in Information Visualizations

“…Both experiments are focused on the automatic visualization generation. The generator use code templates based on a meta-model to define dashboards [2][3][4] and a Python script in which the different parameters are tuned to get a set of visualizations. The script processes the dataset changing a set of characteristics and provide a HTML and JavaScript file with the visualizations.…”