RESEARCH LINES

Acquisition, modeling and knowledge extraction

This line of research is dedicated to the development of new techniques that improve the capabilities of obtaining and interpreting complex data, through the integration of the multiple formats in which they are presented and the standardization of tools to extract information from them.

Existing techniques are not suited to the current data scenario, whose complexity is given by the variety of formats in use, their scale, the speed of change, and the variability in the quality and completeness of the information.

Therefore, we have set out to explore the foundational representations and languages of data with these challenges in mind, in order to study the development of new data search and reasoning techniques adapted to this new scenario.

Our final objective is to generate integral solutions to the problems described above, with emphasis on:

  • Achieve a multi-model approach, which abstracts from the specifics of each particular format by providing an integrated view of the variety of sources available.
  • The development of techniques for managing these targets at scale.
    The design of adaptive notions of approximating query response and reasoning in the face of uncertain or incomplete data, with qualitative or quantitative guarantees, when exact execution is not feasible.
  • The incorporation of statistical operators in query languages that simplify subsequent analyses.

Partner universities