Neural network based intelligent analysis of learners' response for an e-Learning environment

Abstract

This paper presents a novel neural network based scheme for intelligent analysis of learners' response for an e Learning environment. The scheme applies to typed-in single word textual response from the learners' end. The proposed system is intelligently adaptive to inadvertent mistakes committed by the learner while responding to the system's queries. Two-layered feed-forward neural net is employed to build the analyzer. Experimental results, carried out on a wide variety of sample responses show that the proposed scheme successfully simulates a human instructor communicating with the learner through single-word typed-in text.

An adaptive scheme for handling deletion spelling errors' for an intelligent e-learning system

“…Learner; Typed·1 n Response Systems Ded�on This paper augments the proposed scheme for intelligent analysis of leamer's response in an e-learning environment [4] which is based on a cognitive model as shown in Fig. 3.…”

“…In [4], Chakraborty and Roy have proposed an Artificial Neural Network based system which classifies learners' response as correct or incorrect depending on the spelling of the response word. They also proposed a scheme for encoding the input string.…”

“…The work as is [4] takes care of errors of substitution and transposition but not of deletion which would comprise of about 20% of spelling errors. This paper proposes a technique to be employed for response words suffering from errors of deletion.…”

Survey of Techniques Used for Answer Evaluation using Semantic Network

“…Multiple choice questions not only fail to credit students for partial knowledge but also may credit answers even if the learner is in state of absence of knowledge or partial or full misconception because it is also possible to score in such tests using pure guess work [2] The claim of higher efficacy of open ended questions over close ended ones is still debatable. However, what goes without contention is that a wider range of the learners' ability can be tested using open ended questions [3].…”