Connecting domain-specific features to source code: towards the automatization of dashboard generation
- PDF / 1,535,540 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 57 Downloads / 241 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Connecting domain-specific features to source code: towards the automatization of dashboard generation Andrea Va´zquez-Ingelmo1 David Fonseca Escudero4
•
Francisco Jose´ Garcı´a-Pen˜alvo1 • Roberto Thero´n1,2 • Daniel Amo Filva`3
•
Received: 9 September 2019 / Revised: 15 October 2019 / Accepted: 29 October 2019 Ó Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract Dashboards are useful tools for generating knowledge and support decision-making processes, but the extended use of technologies and the increasingly available data asks for user-friendly tools that allow any user profile to exploit their data. Building tailored dashboards for any potential user profile would involve several resources and long development times, taking into account that dashboards can be framed in very different contexts that should be studied during the design processes to provide practical tools. This situation leads to the necessity of searching for methodologies that could accelerate these processes. The software product line paradigm is one recurrent method that can decrease the time-tomarket of products by reusing generic core assets that can be tuned or configured to meet specific requirements. However, although this paradigm can solve issues regarding development times, the configuration of the dashboard is still a complex challenge; users’ goals, datasets, and context must be thoroughly studied to obtain a dashboard that fulfills the users’ necessities and that fosters insight delivery. This paper outlines the benefits and a potential approach to automatically configuring information dashboards by leveraging domain commonalities and code templates. The main goal is to test the functionality of a workflow that can connect external algorithms, such as artificial intelligence algorithms, to infer dashboard features and feed a generator based on the software product line paradigm. Keywords SPL Domain engineering Meta-model Information dashboards Feature model Artificial intelligence Automatic configuration
1 Introduction Nowadays, a lot of technological contexts ask for tailored products; new user profiles have arisen due to the extended use of technologies, and these profiles might demand distinct features in their products. The use of software tools is & Andrea Va´zquez-Ingelmo [email protected] 1
GRIAL Research Group, Computer Science Department, Research Institute for Educational Sciences, University of Salamanca, Salamanca, Spain
2
VisUSAL Research Group, Computer Science Department, University of Salamanca, Salamanca, Spain
3
Departament d’informa`tica, La Salle, Universitat Ramo´n Llull, Barcelona, Spain
4
Departament d’arquitectura, La Salle, Universitat Ramo´n Llull, Barcelona, Spain
not restricted and limited to technical profiles anymore, which have expanded the variety of products that one can find to solve specific problems. This increase in technology usage has also increased the quantity of available and generated data.
Data Loading...
