PhD Alumni

Daniel Florian

Present position: Post-doc researcher at University of Trento

Thesis title:  Model-Driven Design of Context-Aware Web Applications
Advisor:  Stefano Ceri
Research area:  Web, multimedia and databases
Thesis abstract:  
1. Introduction and description of objectives
The evolution of the Information Technology in the last years has seen the World Wide Web transforming from a read-only hypertext media into a full-fledged, multi-channel and multi-service application delivery platform. As a consequence, there has been an evolution from simple, static Web sites to complex, data-intensive Web applications. As for the development of such Web applications, the described evolution demands for appropriate development methods, able to cope with the growing complexity and the specific peculiarities of such new generations of Web applications. It is the field of Web Engineering that addresses this demand and that aims to develop systematic methodologies and solutions for an efficient development process for modern Web applications. With the advent of new and powerful mobile devices, the Web is addressing a continuously growing number of users and is more and more pervading our everyday life. In this regard, the need to improve the user’s browsing experience, e.g., by adapting the application to user preferences and device characteristics, has become manifest. Personalization and adaptation to preferences and device characteristics have already proved their benefits for both application providers and content or service consumers. Similarly, context-awareness and more flexible adaptation mechanisms are increasingly becoming key factors to enhance both the effectiveness and the efficiency of the Web applications of today and especially of tomorrow. “Context-awareness” is intended as capability to take into account whichever properties or information that characterize the interaction with the application, i.e. the context, and to react to changes that such properties or information may experience during the use of the application. Reactions, i.e. application adaptations, are therefore not anymore based on the sole user preferences and device characteristics, but more in general on any property that characterizes the context of the interaction. Typical application adaptations in Web applications are, for example, the adaptation of contents or hyperlinks, the execution of operations or services, or the adaptation of presentation or style properties.

2. Methodology
In line with the previous considerations, this dissertation puts its focus on the development of context-aware and adaptive Web applications. As answer to the challenge faced by the Web Engineering field, this dissertation proposes a conceptual, model-driven method for the design of context-awareness and adaptivity in Web applications. The proposed method is achieved by extending an already established conceptual modeling language for Web application design, i.e. the Web Modeling Language (WebML), also providing for the automatic generation of the application code. The proposed design model reflects a conceptualization of problems and solutions deriving from the use of context-aware and/or adaptive features in the domain of the Web, thus representing a comprehensive instrument, covering the main requirements in the design of context-aware Web applications. This dissertation provides one of the first methodological approaches to context-awareness and adaptivity in the field of Web Engineering. More precisely, this dissertation is one of the first attempts to enlarge the applicability of adaptive application features in the Web from “adaptive hypermedia systems” to “context-aware Web applications”. While the former typically are based on a user model that is dynamically updated based on the observation of the user’s navigation actions, the latter may be based on a more complex context model and active, context-triggered application features. Although the research described in this dissertation is applied to the WebML method, its general nature also contributes to the advancement of the Web Engineering field in general.

3. Discussion of the results obtained
The results that have been achieved in the context of the research described in this dissertation can be summarized as follows:
 Concepts and techniques from the fields of context-aware computing, context modeling, ubiquitous and/or pervasive computing have been applied successfully to the domain of the Web.
 The novel ideas have been appropriately formalized by extending a well-known conceptual modeling language/method for the design of Web applications, i.e. WebML [1,2], keeping the intuitive, visual modeling paradigm that characterizes the language.
 The extended modeling method has yielded the realization of a proper modeling instrument by adapting the visual CASE tool for WebML, WebRatio. The tool is equipped with a powerful automatic code generator that enables the fast and efficient implementation of adaptive Web applications [2]. The strong compliance with WebML and WebRatio maximizes the extensibility of the developed solution.
This dissertation discusses typical issues that may arise when introducing context-awareness to the discipline of Web engineering, and it extends the terminology used to describe adaptive or context-aware Web applications. An analysis of similar modeling approaches shows that there are still subtle differences in the terminology used and, also, in the solutions proposed. We hope that this dissertation also contributes to make a small step forward toward a common terminological framework in the area of adaptive and/or context-aware Web applications.

4. References
[1] Stefano Ceri, Florian Daniel, Maristella Matera, Federico M. Facca. “Model-driven Development of Context-Aware Web Applications”. ACM Transactions on Internet Technology (TOIT), Volume 7, Number 1, February 2007, ACM Press.
[2] Stefano Ceri, Florian Daniel, Federico Facca, Maristella Matera. “Model-driven Engineering of Active Context Awareness”, in World Wide Web Journal, special issue on Multi-channel Adaptive Information Systems on the World Wide Web, ISSN 1386-145X (Print), 1573-1413 (Online), March 2007.