Use this URL to cite or link to this record in EThOS:
Title: Adaptive learning activity selection for an intelligent tutoring system
Author: Alhathli, Manal Ali
ISNI:       0000 0004 7972 5222
Awarding Body: University of Aberdeen
Current Institution: University of Aberdeen
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Providing personalized learning content supports learners to find learning paths that satisfy their needs and ability. This thesis investigates adaptive learning activity selection. Based on the literature, it investigates the impact of learner characteristics such as personality traits, learning styles, cognitive styles, prior knowledge and competence on learning activity selection, and the extent to which learning activities are perceived to be enjoyable and to increase confidence and skills. The adaptation of learning activity selection was investigated in three phases. Phase 1 considered the selection of learning activities that differ in style: active/passive, individual/social, active/reflective, sensory/intuitive, sequential/global, visual/verbal, and modalities related to the cognitive styles verbalizer/visualizer. First, we studied in two learning domains the effect of the Big Five personality traits (in particular extroversion and conscientiousness) on the selection of learning material styles (active/passive, individual/social), and the extent to which learning material styles are perceived to be enjoyable and to increase confidence and skills. We found a small effect of personality. Second, we studied the impact of learner personality (extroversion and openness to experience) and learning styles on the selection and appreciation of learning material styles (active/reflective, sensory/intuitive, sequential/global, visual/verbal). We found little evidence that learning styles matter. Third, we studied the impact of learners' verbalizer/visualizer cognitive styles on the selection and appreciation of learning material styles. We found limited evidence that cognitive styles matter. Phase 2 considered the selection of learning activities that differ in knowledge complexity: the type and amount of required prior knowledge and knowledge taught. First, through a qualitative focus group study, we investigated the impact of personality traits (openness to experience, emotional stability, and self-esteem) on the selection of the next learning activity, taking into account the learning activities' use of prior knowledge and topics taught. We found a clear impact of all three traits on the selected activities, with different activities being selected for learners high in the traits than for learners low in the traits. Personality impacted both the required prior knowledge and the knowledge taught. The focus groups also indicated that learner competence matters when making decisions on activity selection. Next, through an empirical study, we investigated the impact on learning activity selection of learners' emotional stability and competence. We found that emotional stability and learner competence both impacted the learning activities selected for learners. In Phase 3, based on the data from Phase 2, we propose three algorithms that select learning activities with varying required prior knowledge and taught knowledge adapted to learner emotional stability and competence. Finally, through a study with teachers, we evaluate our algorithms and propose a final one.
Supervisor: Masthoff, Judith ; Beacham, Nigel Sponsor: Ministry of Education Saudi Arabia
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Intelligent tutoring systems ; Learning ; Recommender systems (Information filtering) ; Personality