姓名:謝宏偉 學號: M99G0219 班級:碩研資工一甲 201O 2nd International Conference on Education Technology and Computer ( ICETC) Neural Network Based Intelligent Analysis of Learners' Response for an e-Learning Environment
Outline 1. INTRODUCTION 2. LEARNERS RESPONSE IN AN E-LEARNING ENVIRONMENT 3. INTELLIGENT RESPONSE ANALYSIS 4. PROPOSED SCHEME 5. EXPERIMENTAL RESULTS 6. CONCLUSIONS
1. INTRODUCTION 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.
2. LEARNERS RESPONSE IN AN E- LEARNING ENVIRONMENT Typed interactions may take place in an e- Learning system under two circumstances, viz., dialog between the learner and the system, to simulate the real-life learning experience and assessment of learning achievement. While objective type interaction is close ended and guided, based on the options available in the test item itself, dialog based test items need embedded intelligence to assess.
2. LEARNERS RESPONSE IN AN E- LEARNING ENVIRONMENT To provide a more meaningful learning experience, or assessment of learning achievement, under an E-learning environment, a learner must be offered the scope of interacting with the system through typed-in texts.
3. INTELLIGENT RESPONSE ANALYSIS The limitations of MCQ's are covered to some extent by close ended questions having text based answers with single words or a few sentences in which case we get the best of both worlds.
3. INTELLIGENT RESPONSE ANALYSIS
4. PROPOSED SCHEME Neural Networks have been trained and used as effective classification tools where the values in the output neurons are indicative of the class to which the input patterns culminate. Viable and successful applications of ANN's as language classifiers are available both for text and phonetics based approaches.
4. PROPOSED SCHEME
5. EXPERIMENTAL RESULTS
6. CONCLUSIONS The proposed scheme successfully simulates a human instructor communicating with the learner through single-word typed in text. Experimental results show that the system is intelligent enough to simulate the interaction between a human instructor and a learner when the learner's response is restricted to single word.