In this topic, our research is related to the development and evaluation of new paradigms and tools for conceptual modeling of both traditional and non-traditional applications. A tool recently developed as part of a work in this topic is OMT-G, an extended version of the OMT model for geographical applications that has been adopted by several institutions in Brazil.
Cooperative interfaces are interfaces that assist the user to formulate queries to semistructured and structured dataases. Our activities in this research topic aim at the development of new query paradigms and the implementation and evaluation of interfaces based on these paradigms.
Our research activities in this topic focus on the development of methods and tools for database design, mainly directed to the use of the ER and object-oriented models for conceptual modelling and generation of (relational) database schemas that correctly represent the semantics of the corresponding application.
In cooperation with the Digital Libraries Research Laboratory at Virginia Tech, our research work in this topic aims at incorporating in information services for digital libraries features such as: seamless interoperability of data and text repositories; consistent fusion of results from search over distinct heterogeneous information sources; use of several types of evidences to assess the relevance of retrieved objects in a query result; and new paradigms for users to issue queries and to analise their results. This joint work is developed in the context of the Networked Digital Library of Theses and Dissertations (NDLTD).
Web and Semistructured Data Management
In the last few years, the group has been involved with the development of methods and tools for dealing with data available in the Web and in other non-structured medium (e.g., documents). The main emphasis has been in providing facilities similar to those available in traditional database systems for managing such data. This research work is mainly characterized by data extraction tool we have developed, called DEByE, and a number of companion tools for manipulating and querying semistructured data available in Web sources.
Web Information RetrievalIn this research topic, we are interested in the development of information retrieval technology for dealing with the huge amount of information available on the World Wide Web. Our research includes techniques for generic and focused Web crawling, and a model for automatic categorization of medical documents. This particular model is now being extended for dealing with other controlled domains such as the law domain. The results of this research were used in two commercial country wide search engines (TodoBR and TodoCL) as well as in medScopio, a search engine specialized in the