|
Invited Lecturers
 |
|
Dr. Jose Luis Verdegay (More Information: Curriculum, Web)
Professor of Computer Science and Artifical Intelligence University
of Granada (UGR).
Conference:
The Pillars of Soft Computing
In the area of Intelligent Systems, the Soft Computing is a methodology that has repeatedly demonstrated its efficiency and effectiveness of face to model situations and solve complex problems in a wide range of areas, but there are still technological environments in which there has tested their applicability or, if attempted, have been poorly studied its possibilities. This may be the case of Renewable Energy Assessment Service Quality, Technology Foresight, logistics, systems biology and others. This advanced technology environments of crucial importance to our society because of the importance that may have solutions that are given to the problems which they appear, may affect our way of life and relationship.
Since 1965 L.A. Zadeh gave the fuzzy set definition, applications and developments based on this simple concept has evolved so that today it is virtually impossible to calculate the volume of business generated worldwide and can be found whose operation is Intelligent Systems directly based on this concept, ranging from the most everyday drivers, even the most sophisticated models for system identification. But while the historical trace sets and fuzzy systems is well known, it's the same with that of the Soft Computing, despite its current popularity, so it is essential to put black on white origin and definition.
The establishment of what we call the pillars of Soft Computing, will be the central argument of this conference, which also show the potential application that has the Soft Computing, specifying a particular field and particularly describing how to focus correctly with this method the solving different problems of different nature, thereby facilitating the subsequent design, production and development of new Sustainable Intelligent Systems, ie, which by their nature are not doomed to stay in the realm of "theoretical", but rather the opposite: that by being equipped with enough arguments and resources, are intended to be transferred safely to the productive sector.
|
 |
|
Dr. Tom M. Mitchell (More information: Curriculum, Web)
Professor Carnegie Mellon University, Machine Learning Department.
Conference:
Forever Learning
This talk considers the question of how to construct intelligent software agents learn That forever.We present one case study - a system NELL That Has Been Called daily learning for well over a year, 24 hours Each Day. NELL Goals Each day has two. The first is to read, or extract, more information from the web to populate STI growing knowledge base of structured knowledge.The second is to learn to read Better Than yesterday, as evidences by ITS Ability to go back to the Same it read web pages yesterday, and extract more facts more accurately today. The talk will describe the design of NELL, STI Successes and Its Failures (so far), and use this case study to explore more braodly the quesiton of how to design machine learners never-ending. Puede follow NELL's progress daily at http://rtw.ml.cmu.edu/.
|
|
|
Dr. Maria-Amparo Vila Miranda (More information: Curriculum, Web)
Professor of Computer Science and Artificial Intelligence University of Granada.
Conference:
The treatment of imprecise information and uncertain, some solutions and new problems.
Few scientific communities in the field of Computer Science have thought much about their present and future as researchers engaged in databases. For over fifteen years a very large group of renowned authors in this area meet regularly to fix the main challenges and problems to solve than anticipated, and suggest, based on this line of research necessary and promising. His conclusions were reflected in a set of reports that have appeared in publications of ACM and SIGMOD and have been widely disseminated.
One problem that appears most consistently and strongly in these reports is the treatment of imprecise and uncertain information in databases. Already in 1991 is cited as one of the factors to be taken in the design of a new data model. Subsequently, in each and every one of his reports continued to raise this target as a priority, both from the point of view of access to databases and from the Data Mining.
Treatment by Fuzzy Logic of imprecise and uncertain information in databases is a well established line of research that has been in existence for over 20 years. The use of Fuzzy Data Mining problems are also widespread. In fact even before this area of work is configured as such, were treated the same problems inherent in using fuzzy approaches. Subsequently we have worked to combine the power of representation and manipulation of imprecise knowledge of fuzzy logic provides the capabilities that have the Neural Networks and Genetic Algorithms and Bioinspired, providing an excellent approach to data mining.
The aim of this talk is to present how the Soft Computing and Fuzzy Logic particular have addressed the solution of important problems imprecision and uncertainty in information systems, collecting also the results obtained by our research group and the applications we have developed.
|
|
|
|
|
|
|
|
Invited Lecturers
