Dr. AndreaBaruzzo

Academic homepage

Research Projects

In this page you can find more information about the research project I am currently involved on.

  • UMLaid, an automated framework for UML model analysis;
  • E-Dvara 2, a service-oriented digtal library for EU-India cultural heritage;
  • PIRATES, a framework for personalized, intelligent recommendations and annotations of Web documents



UMLaid is a tool set that is designed from the beginning to pervasively supports the activities discussed in the previous section. More in particular, we have conceived the UMLaid framework to be structured in order to support the following V&V procedures:

  1. Validation of a UML model and its textual specification through the support provided by static checkers and model animators/simulators;
  2. Verification for consistency and completeness of the UML model with respect to the associated OCL specification;
  3. Forward and reverse engineering aimed to synthesize and synchronize models and code in a MDA setting;
  4. Model-based testing activities;
  5. Evaluation of the overall design quality aimed to identifying potential design flaws), especially with respect to good design principles (design patterns, quality metrics and design heuristics).
Some demonstrators have been developed in Haskell as proof-of-concept tools. Currently, some of these tools are under development as a plug-in of commercial software tools (such as Enterprise Architect).


E-Dvara 2

E-Dvara is a project focused on the development of a new platform for storage of digital contents. Since its inception, it has been explicitly designed to overcome some limitations that characterize the process of building a digital content. In particular, E-Dvara was initially meant to:

  1. reduce the effort required by the archivist to define the data structure used to represent data into the archives;
  2. provide to archivists with no expertise in data management a set of wizards devoted to data schemata creation in a completely automatic and transparent way (with respect to the physical database);
  3. allow content providers to easily share their archives on the Web by means of a build-in Web interface or with several other applications, allowing archivists and system administrators to define the way data should be rendered to final users;
  4. allow archivists to provide for each archive of digital contents a specific visualization template and a set of search forms.
More recently, we started the integration in the E-Dvara platform of a set of semantic services aimed at assisting final users to select from the archives the most appropriate content with respect to their current information needs. This new preototype is a complete new system, called E-Dvara 2, which replaces the old prototype and represents our first step in the direction of realizing a semantic digital library.



PIRATES (Personalized Intelligent Recommender and Annotator TEStbed) is a general framework for text-based content retrieval and categorization which exploits social tagging, user modeling, and information extraction tech- niques. Using an integrated set of tools, this framework lets the users experiment, customize, and personalize the way they retrieve, filter, and organize the large amount of information available on the Web. Furthermore, the PIRATES framework undertakes a novel approach that automatizes typical manual tasks such as content annotation and tagging, by means of personalized tags recommendations and other forms of textual annotations (e.g. key-phrases).