Knowledge Engineering Group (KEG)
KEG is a research group devoted to the following main research areas:
- Knowledge Representation and Reasoning
- Knowledge Discovery and Embedded Analytics
- Health and Biomedical Informatics
Knowledge Representation and Reasoning is complemented with soft computing approaches to problem solving that caters for the handling of incomplete, unknown, or even contradictory information. This development has led to a situation where we, in a manner of speaking, are at risk of drowning in the overwhelming flow of information. To be able to extract the relations that are pertinent in a given situation and to present them in a way that it is simple to understand, is rapidly becoming the main problem. Efficient solutions to this problem will be crucial components in future application arenas where evolutionary computation plays an essential role, leading to quantic intellects.
Knowledge Discovery and Embedded Analytics is a set of ongoing activities that enable the extraction and analysis of useful knowledge. The main goal is to discover useful, valid, relevant and new knowledge about a particular activity through algorithms, taking into account the magnitudes of data increasing. The goal of the research in this particular field has been to present valid and sound solutions to particular problems, mostly in real or critical world. Solutions oriented to monitoring and reporting has been addressed and issues related to knowledge discovery using data mining or soft computing, in particular evolutionary computation, will be developed for advanced knowledge management and analysis.
The overall purpose of systems developed in the field of health and biomedical informatics is to extract and present clinically relevant information. Biomedical information appears in many different forms, i.e., parameters, value measurements, time courses, images, volumes, and image and volume sequences. Methods to attain new types of information are continuously being developed and the detail and quality of recorded data is increasing rapidly. Huge amounts of potentially relevant information can be tied to one single patient. In addition, it is necessary to be able to integrate and analyze information from a large number of patients and time instances. To develop principles and methods for such solutions is the goal of the research in the field of health informatics in the next years. In particular, the research group focuses on the development of systems for biomedical decision support, in particular in healthcare. The result of these systems tends to be the improvement of human experience and quality of life without an explicit perception of the underlying interaction between technology.
KEG leader: José Machado