| dc.description.abstract | This chapter describes the development of an expert system by a knowledge engineer with pedagogical expertise. Specifically, the problem of knowledge elicitation from the expert's point of view is thoroughly presented as characterized by skillful coding in the expert's system input and output representation of knowledge. Based on the developed expert system, neural networks and neurofuzzy logic technologies can be used in order to achieve automatic evaluation of deaf students' answers and progress. The students have basic knowledge of the English language and computer skills in a virtual e-learning environment. Using the expert system's knowledge representation, different backpropagation neural (BPN) models can be developed and evaluated using the correlation coefficient between the values of the neural network's response and the data values, plus the count in percentage of the error between the prediction values of the neural network and the real values of the data during training and afterward in unknown data that haven't been used during learning. Also using the expert system's knowledge representation, different adaptive neurofuzzy inference system (ANFIS) models can be developed and evaluated using the count in percentage of the error between the expected values and those provided by the ANFIS models. In addition, general description for the characteristics of the neural networks and neurofuzzy technologies is provided along with their use in similar problems. © 2010 Springer-Verlag US. | en |
