Objectives Evaluation of existing models, languages, and techniques for information integration. The appraisal of the state of the art in this field will encompass theoretical research (e.g. on database transformations, the problem of answering queries using views, and data integration based on description logics and logic programming) as well as research systems and commercial systems, both in the context of data warehousing and more speculative architectures.

Description of work TASK 1: Appraisal of models and systems for information integration. (Task Leader UNIROME1) The problem of information integration is currently addressed in several different fields, e.g. in database research, software engineering, and artificial intelligence. Recent proposals concerning information integration in all these fields will be analyzed. Since providing a comprehensive framework for information integration is one of the main project goals, this activity aims at identifying, among all the relevant aspects that a comprehensive information integration system should address, which aspects are actually taken into account by state-of-the-art methods and systems, and which aspects are not. Moreover, the relationships among the various approaches to information integration will be highlighted. TASK 2: Appraisal of methods and systems from computational logic. (Task Leader UNICAL) A key idea of the INFOMIX project is the exploitation of Computational Logic for Information Integration. To this end, the task will make an appraisal of the main methods and systems which are nowadays available in the area of computational logic, analyzing their features and their suitability for information integration tasks. TASK 3: Appraisal of techniques and systems for the acquisition and transformation of heterogeneous data. (Task Leader TUWIEN) Structural issues in data integration (often approached with the popular wrapper metaphor) need to be covered in every complete data integration architecture, with INFOMIX being no exception. This task will consider data translation and wrapping from the distinct viewpoints of the artificial intelligence and the database communities, both in the contexts of Web integration and the integration of traditional legacy systems. Extraction of information which is not explicitly represented will also be addressed. Part of this task will be the evaluation of available software for the data acquisition and transformation tasks and their possible incorporation into an information integration architecture.

Deliverables D1.1- Review of models and systems for information integration (Report). This report will present a survey of state-of-the-art methods and systems for information integration. D1.2- Review of methods and systems from computational logic (Report). The report will provide a comparative overview of the main computational logic systems from the perspective of information integration. D1.3- Review of techniques and systems for the acquisition and transformation of heterogeneous data (Report). The report will give an overview of approaches to structural data integration previously used both in research prototypes as well as in commercial systems.

Milestones and expected results Selections of the principal research results to be applied and extended within the INFOMIX project. Selections of standards and the software tools to provide the technological basis for the INFOMIX architecture.