fiee 15(2): e3

Research Article

Evaluating an Intelligent Q&A System for Mobile Cultural Learning

Download889 downloads
  • @ARTICLE{10.4108/fiee.1.2.e3,
        author={Ioannis Doumanis and Serengul Smith},
        title={Evaluating an Intelligent Q\&A System for Mobile Cultural Learning},
        journal={EAI Endorsed Transactions on Future Intelligent Educational Environments},
        volume={1},
        number={2},
        publisher={ICST},
        journal_a={FIEE},
        year={2015},
        month={6},
        keywords={assisted learning, intelligent interaction, pervasive \& ubiquitous education, technology supported education, cultural information},
        doi={10.4108/fiee.1.2.e3}
    }
    
  • Ioannis Doumanis
    Serengul Smith
    Year: 2015
    Evaluating an Intelligent Q&A System for Mobile Cultural Learning
    FIEE
    ICST
    DOI: 10.4108/fiee.1.2.e3
Ioannis Doumanis1,*, Serengul Smith2
  • 1: CTVC, 9-10 Copper Row, Tower Bridge Piazza, SE1 2LH, London
  • 2: Research Group on Development of Intelligent Environments Department of Computer Science, Middlesex University London
*Contact email: ioannis@truetube.co.uk

Abstract

The paper presents a user study designed to evaluate the impact of an intelligent Question and Answering (Q&A) system for mobile cultural learning on the user’s subjective impressions and retention performance. We have chosen a recent pedagogical framework for mobile learning as a theoretical foundation for this work. The framework postulates four types of mobile learning from which we have chosen to implement the second one (i.e., High Transactional Distance and Individualised Mobile Learning (HI))[13]. In particular, our Q&A system enables individual learners to interact with unknown cultural content under simulated mobile conditions in a well-organised and structured manner using natural language. To investigate the impact of the Q&A system on the participants’ retention of cultural content, we compared two variations of the Q&A system in the lab under simulated mobile conditions. The systems differed both in terms of the approach to processing natural language (i.e., scripts vs. linguistic parsing) and style of a Q&A session (free vs. constrained). Below, we present the results of the study and a series of design recommendations that should aid in the development of more robust Q&A models for mobile cultural systems.