Domain engineering for generating dashboards to analyze employment and employability in the academic context

Domain engineering for generating dashboards to
analyze employment and employability in the
academic context
Andrea Vázquez-Ingelmo, Francisco J. García-Peñalvo, Roberto Therón
GRIAL Research Group,
Research Institute for Educational Sciences,
University of Salamanca, Salamanca, Spain
andreavazquez@usal.es, fgarcia@usal.es, theron@usal.es

Index
1. Introduction
2. Methodology
3. Results
4. Discussion
5. Conclusions
GRIAL Research Group - TEEM 2018

Introduction
• Informed decision-making processes have gained relevance given
the potential benefits of using data to support strategies
• Necessity of powerful tools for managing and analyzing large
volumes of data to support knowledge generation
• Information dashboards are one of the most commonly used
software tools to explore data in an interactive and friendly way

Introduction
• There are complex fields of study, like employability, that could
take advantage from these tools
• The employability concept has not yet a strong theoretical
foundation given the complexity of discovering influential
indicators to asses it

Introduction
• However developing information dashboards is not a trivial
task
• Several requirements can be involved and can vary among the
different user profiles

Introduction
• Software engineering paradigms like software product lines (SPL)
provide solutions for managing sets of differing requirements,
focusing on the reutilization and composition of base software
assets
• SPL improve:
o Scalability
o Maintainability
o Customization
o Development time

Introduction
• Experimental framework for generating customized dashboards
applying the SPL paradigm
• Case study on the Spanish Observatory for University
Employment and Employability
o Organization with the vision to become an information
reference for understanding the behavior of the variables
related to employment and employability of students from
Spanish universities

Methodology
Domain engineering
• The domain of the product line to be developed is studied in order
to identify the commonalities and variability points of the family
of products
• By identifying these characteristics, it is possible to model the SPL
through a feature diagram

Methodology
• Three components have been modeled for this case study
• These components are three kind of information visualizations that
could have different data sources, functionalities or even layout

Methodology
Code generation
• The identified components were implemented through code
templates
• Variability points (features) were implemented through macros
• The configuration phase is automated through a template based
code generator

Methodology
DSL
• A DSL was designed to abstract the product configuration process
• The features of the product to be generated are specified through
XML technology
• The DSL and code templates feed a code generator that
configures the product accordingly

Methodology
DSL

Methodology
Code generator

Results

Discussion
• The software product line paradigm helps to manage different
and dynamic requirements by providing a theoretical framework
for implementing modular, configurable and reusable software
components
• It is important to take into account the heterogeneity of the
sources involved to decouple the logic of the software
components from the information that they will finally present

Discussion
• Software interfaces require both the study of the domain in which
they will be framed and the study of the target users that will end
up using the products
• It is important to provide not only aesthetic but also functional
and helpful dashboards
• Automatic generation of highly customizable and usable
interfaces is still a tough process that require semi-automatic or
even manual design processes

Conclusions
• An automatic method for generating customized dashboards to
analyze the organization’s data regarding university employment
and employability has been provided
• Developing a framework to efficiently generate flexible and
customizable information dashboards could give a strong
foundation to create powerful tools with the main goal of helping
decision-makers to take well-informed decisions

Discussion

Domain engineering for generating dashboards to analyze employment and employability in the academic context

Recommandé

Recommandé

Contenu connexe

Similaire à Domain engineering for generating dashboards to analyze employment and employability in the academic context

Similaire à Domain engineering for generating dashboards to analyze employment and employability in the academic context (20)

Plus de Technological Ecosystems for Enhancing Multiculturality

Plus de Technological Ecosystems for Enhancing Multiculturality (20)

Dernier

Dernier (20)

Domain engineering for generating dashboards to analyze employment and employability in the academic context