COPAL — Cognitive personalized aid for learning
Authors:
- 1 March 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
Concept learning can be illustrated based on a layered model of knowledge discovery. To discover association between learners' requirements and learning materials we propose a cognitive learning architecture. Development of cognitive personalized aid for learning (COPAL) involves utilizing the concepts of cognitive informatics for providing peer interaction between learner and system and to deal with the challenge of handling unstructured Big E-learning data. Cognitive learning system utilizes the concept of data mining and natural language processing. We present a learner centered strategy of learning using open source Hadoop cluster. The paper describes development of system in which the cognitive learning engine interacts with the learner and the personalized recommender to satisfy the learner needs. Our system provides a scalable collaborative framework for constructing recommendations using extensible library of data mining provided by Apache Mahout. HDFS has been employed for reliability, scalability and low cost storage capability. During data preprocessing Pig has been used to transform unstructured Big E-learning data. The paper demonstrates the use of Pearson correlation and Euclidean distance for similarity computation. Our implementation found it beneficial to use Euclidean similarity metric. It is thus feasible to develop efficient E-learning recommender system coupled with cognitive framework using open source software.Keywords
This publication has 15 references indexed in Scilit:
- DESIGN AND DEVELOPMENT OF A RECOMMENDER SYSTEM FOR E-LEARNING MODULESJournal of Computer Science, 2014
- Watson: The Jeopardy! Challenge and beyond2013
- An Evaluation of Cassandra for Hadoop2013
- Interpreting Concept Learning in Cognitive Informatics and Granular ComputingIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2009
- Pig latin2008
- Some useful tactics to modify, map and mine data from intelligent tutorsNatural Language Engineering, 2006
- Data mining in the e‐learning domainCampus-Wide Information Systems, 2004
- Building a recommender agent for e-learning systems2003
- A step toward the foundations of data mining2003
- Classtalk: A classroom communication system for active learningJournal of Computing in Higher Education, 1996