Using Genetic Programming & Neural Networks for learner evaluation JOHN VRETTAROS 1 , JOHN PAVLOPOULOS 1 , GEORGE VOUROS 2 , ATHANASIOS S. DRIGAS 1 1 NCSR DEMOKRITOS, Institute of Informatics & Telecommunications, Net Media Lab Patriarhou Grigoriou, 15310 Ag. Paraskevi, Greece 2 Aegean University, Info and Communication Systems Eng, 83200, Karlovassi, Samos, Greece Abstract: - The goal of this study is the development of an assessment system, with the support of a Neural Network approach optimized with the use of Genetic Programming. The data used as training data, are real data derived from an educational project. The system developed, is proved capable to assess data from both single select and multiple choice questions in an e-learning environment. The final result is the assessment of the learners’ answers in various criterions. Key-Words: - Neural Networks, genetic programming, assessment, learners 1 Introduction This paper presents the development of an assessment system of the gained knowledge of students. In specific, the results of the self- assessment exercises provided by a learning environment are examined, in order for the students to obtain the knowledge level they have possessed, in each learning section solely and overall. The final aim is the assessment system to be trained in order to play the role of an instructor. The assessment system is based on a neural network approach, optimized with the aid of Genetic Programming. Genetic Programming provides a way to develop a computer program that produces some desired output when presented with particular inputs. Many seemingly different problems as artificial intelligence, symbolic processing and machine learning can be viewed as capable for the development of such a computer program [3]. One of these problems is neural network design and training for data classification [7]. In this paper, in order to produce the assessment system acquired, we examined if the Genetic Programming Neural Network (GPNN) approach [4,5] is able to model the assessment role of a pedagogical expert through data classification. GPNN Assessment System (GPNNAS) is a GPNN system that is trained with data, consisted of answers of students and their evaluation according to a pedagogical expert. GPNNAS should be able to evaluate the answer of a student according to some criteria. The final system consists of one Neural Network for each criterion, optimized with a Genetic Programming approach, so that each neural network is able to evaluate the answer according to the specific criterion. The output of this assessment system is an evaluation of a student’s answer for each criterion. The data generated by the learner going through a mini-test consists of a string of characters and values, which are built, based on certain criteria and have a different pattern depending on the question type. The types of questions are single-select type questions, and multi-select type questions, which have several answer options and any number of them can be selected. The questions test learners against more than one sectors and each question has a relevance value against every sector. 7th WSEAS Int. Conf. on ARTIFICIAL INTELLIGENCE, KNOWLEDGE ENGINEERING and DATA BASES (AIKED'08), University of Cambridge, UK, Feb 20-22, 2008 ISSN: 1790-5109 Page 160 ISBN: 978-960-6766-41-1