Home // ICSEA 2022, The Seventeenth International Conference on Software Engineering Advances // View article
Authors:
Najla Maalaoui
Raoudha Beltaifa
Lamia Labed Jilani
Keywords: Service-Oriented Dynamic Software Product Lines; Ontology; User requirements; Natural language processing.
Abstract:
A Service-Oriented Dynamic Software Product Line (SO-DSPL) is a family of service-oriented systems sharing a set of common features. Hence, they are automatically activated and deactivated depending on the running situation. Such product lines are designed to support their self-adaptation to new contexts and requirements. Particularly, user requirements can be analyzed and enriched thanks to the existing of the SO-DSPL ontology that we previously built. This will facilitate the configuration of a derived service from the family of services, corresponding both to the desired requirement and a specific context. As we know, a user requirement can be ambiguous, vague and incomplete, which motivate the need for the extraction of the hidden knowledge. Our challenge is to use artificial intelligence techniques to automatically extract new SO-DSPL knowledge from textual user requirements and derive appropriate services of the service line for the user. In this paper, our approach is based on Natural Language Processing (NLP) learning techniques, a rule engine and a reasoner. This process permits to better understand the user requirements, to predict other information about the requirements and to derive an appropriate service (software application as a combination of several services) in the SO-DSPL application engineering phase. We use the Smart Home product line and a dataset of textual user requirements to evaluate our proposal. Notes that when we say product, we mean an application based on service compositions.
Pages: 58 to 68
Copyright: Copyright (c) IARIA, 2022
Publication date: October 16, 2022
Published in: conference
ISSN: 2308-4235
ISBN: 978-1-61208-997-3
Location: Lisbon, Portugal
Dates: from October 16, 2022 to October 20, 2022