Chapter

Knowledge-Based Intelligent Information and Engineering Systems

Volume 4253 of the series Lecture Notes in Computer Science pp 1012-1019

Neural Network Approach for Learning of the World Structure by Cognitive Agents

  • Agnieszka PieczyńskaAffiliated withInstitute of Information Science and Engineering, Wrocław University of Technology
  • , Jarosław DrapałaAffiliated withInstitute of Information Science and Engineering, Wrocław University of Technology

* Final gross prices may vary according to local VAT.

Get Access

Abstract

In this work an original method for coping with agents’ incomplete knowledge is introduced. This method called the algorithm for the messages generation is applied by the cognitive agents when the states of external objects can not be directly perceived. To approximate the current states of objects all agent’s experience as temporal data base is taken into account. As a result of the algorithm the logic formulas with modal operators are generated. One of the steps of proposed algorithm is the classification of the observations. It is shown how neural network approach might be used in order to determined some tendencies to occurrence specific states of objects.