Bouchra El Asri

Multiview Components for User-Aware Web Services

Component based software (CBS) intends to meet the need of reusability and productivity. Web service technology leads to systems interoperability. This work addresses the development of CBS using web services technology. Undeniably, web service may interact with several types of service clients. The central problem is, therefore, how to handle the multidimensional aspect of service clients’ needs and requirements. To tackle this problem, we propose the concept of multiview component as a first class modelling entity that allows the capture of the various needs of service clients by separating their functional concerns. In this paper, we propose a model driven approach for the development of user-aware web services on the basis of the multiview component concept. So, we describe how multiview component based PIM are transformed into two PSMs for the purpose of the automatic generation of both the user-aware web services description and implementation. We specify transformations as a collection of transformation rules implemented using ATL as a model transformation language.

A model driven approach for context-aware services development

Nowadays, context-awareness constitutes an essential aspect for service oriented applications development. Indeed, applications are supposed to react to context changes in order to provide end-users with pertinent services. To deal with such requirement, Context-Aware Service (CAS) has emerged as an underlying software engineering paradigm for context-aware applications development. In this paper, we present a model driven approach that highlights the fundamental context-awareness challenges for CAS development. So, we first present our context metamodel based on a practical specification of the context. Then, we propose a CAS specification and metamodel and show how can Aspect Paradigm (AP) be used to fulfill service adaptation to its execution contexts. We illustrate our approach by an E-tourism motivating scenario.

Engineering adaptable service oriented systems: A model driven approach

Service Oriented Architecture (SOA) is an emerging architectural style for the development of interoperable, agile, large scale and distributed software systems. Such systems are based mainly on services as core building blocks which may interact with several types of service clients. Hence, the central problem is how to deal with the challenge of adaptability of such services. The present paper proposes a Model-Driven engineering approach for adaptable service oriented systems. The proposed MDD approach consists of (1) a Unified Modeling Language (UML) profile to graphically model adaptable service oriented systems regardless of standards and implementation platforms. The key element of the proposed profile is the multiview service defined as a first class modeling entity that allows the representation of the needs and requirements of users by separating their concerns early in the development lifecycle of service oriented systems (2) an MDD tool that accepts a UML model defined with the proposed profile and transforms it to Java code which constitutes the implementation of each service according to the J2EE platform

Detecting feature duplication in natural language specifications when evolving software product lines

Software product lines are dynamic systems that need to evolve continuously to meet new customer requirements. This evolution impacts both the core platform of the product line and its derived products. For several reasons, the most common way to express requirements by customers is natural language. However, the experience has shown that this communication channel does not give the possibility to detect system defects such as inconsistency and duplication. The objective of this paper is to propose a method to transform textual requirements into the XML format used by some Feature-oriented software development tools, in order to facilitate the detection of features duplication.

Analyzing Trends in Software Product Lines Evolution Using aCladistics Based Approach

A software product line is a complex system the aim of which is to provide a platform dedicated to large reuse. It necessitates a great investment. Thus, its ability to cope with customers’ ever-changing requirements is among its key success factors. Great effort has been made to deal with the software product line evolution. In our previous works, we carried out a classification of these works to provide an overview of the used techniques. We also identified the following key challenges of software product lines evolution: the ability to predict future changes, the ability to define the impact of a change easily and the improvement in understanding the change. We have already tackled the second and the third challenges. The objective of this paper is to deal with the first challenge. We use the cladistics classification which was used in biology to understand the evolution of organisms sharing the same ancestor and their process of descent at the aim of predicting their future changes. By analogy, we consider a population of applications for media management on mobile devices derived from the same platform and we use cladistics to construct their evolutionary tree. We conducted an analysis to show how to identify the evolution trends of the case study products and to predict future changes.

Co-evolution analysis for software product lines

The purpose of our approach is to study the co-evolution of the platform and the products of software product lines. Because the platform must be able to derive all the family products, products are not allowed to evolve independently from the plateform, thus the propagation of the products changes must be managed efficiently. Instead of focusing on the change impact analysis we propose an approach to compare the evolution histories of the products and the platform illustrated through evolutionary trees built using the biological technique cladistics. This comparison yields important results concerning the change propagation. In this paper, we introduce the use of cladistics for software product lines to build evolutionary trees for platform and products, then we elaborate a mathematical analysis to compare these trees, afterwards we validate this work through a case study (mobile media software product lines). We also provide the design of an automated tool.

Return on Investment of Software Product Line Traceability in the Short, Mid and Long Term

Several works discuss tracing in Software Product Lines from a technical and architectural points of view, by proposing methods to implement traceability in the system. However, before discussing this field of traceability, we first need to prove the profitability of integrating such approach in the Product Line. Therefore, we bring in this paper a quantitative analysis on how traceability can impact the Return on Investment of a Software Product Line, and in which conditions, in terms of number of products and SPL phase, can tracing be profitable. We compare the results of a generic Software Product Line estimation model, COPLIMO, and our model METra-SPL. Our analysis shows that introducing traceability costs when constructing the Product Line, but can be profit making in the long term, especially in maintenance phase, starting from 2 products to generate.

Cloud environment assignment: A context-aware and Dynamic Software Product Lines-based approach

With the increase in number of mobile devices deployed in cloud computing, the demand of context-aware services to assign increases. Indeed, Information about the users environment exposes new challenges to cloud computing in terms of location-aware, time-aware, device-aware and personalized applications to cope with the constraints of mobile devices in matters of interaction abilities and communication restrictions. In addition, the user also needs context information about services provided by the provider. For instance, the user can check the availability of service, the response time of service, the cost, quality of the service and quality of context information. This paper proposes a software framework which supports context-awareness behavior to assign services to Consumers and especially mobile users. This framework is based on Dynamic Software Product Line approach to handle this variability and adaptation in context at runtime. In fact, changes can occur in the application context requiring the cloud environment to be reconfigured, for instance, non-functional requirements like response-time, availability or pricing are violated.

Vers un profil UML pour la conception de composants multivues

This paper aims to present an UML profile based on multiviews components. A multiviews component allows to encapsulate and deliver information according to the users point of view and offers mechanisms to manage the dynamic evolution of viewpoints and consistency among views. In this paper, we first present the notion of multiviews component and its integration into a UML profile, and thus detail an object-oriented multiviews modelling applied to a concrete example. The transition to the coding stage is described through a generic implementation pattern.

Using Unsupervised Machine Learning for Data Quality. Application to Financial Governmental Data Integration.

Data quality, means, that data are correct, reliable, accurate and valid to be used and to serve its purpose in a given context. Data quality is crucial to make right decisions and reports in every organization. However, huge volume of data produced by organizations or redundant and heterogeneous data integration make manual methods of data quality control difficult, for that using intelligent technologies like Machine Learning is essential to ensure data quality across the organization. In this paper, we present an unsupervised learning approach that aims to match similar names and group them in same cluster to correct data therefore ensure data quality. Our approach is validated in the context of financial data quality of taxpayers using scikit learn the machine learning library for the Python programming language.