Academic journal article Educational Technology & Society

An Innovative Approach to Scheme Learning Map Considering Tradeoff Multiple Objectives

Academic journal article Educational Technology & Society

An Innovative Approach to Scheme Learning Map Considering Tradeoff Multiple Objectives

Article excerpt

Introduction

Personalized learning has been a popular trend in the field of e-learning. Owing to learners have different learning abilities, knowledge, and learning performance (Tan, Shen, & Wang, 2012), personalized learning aims to fit the needs, goals, talents, and interests for individual learners (Klasnja-Milicevic, Vesin, Ivanovic, & Budimac, 2011). To date, many researchers have carried on studying personalized learning with advanced computer technologies, for examples, Hsu (2008), and Hsieh, Wang, Su, and Lee (2012) have developed the English learning recommender systems to provide learners with appropriate materials according to their profiles and thus help them better English abilities and learning motivations. Moreover, Hwang, Sung, Hung, and Huang (2013) have indicated that learners having learning style-fit materials better learning achievements significantly.

With the rapidly grown Internet, e-learning have explosive learning resources and materials, the customized learning has become an important issue in personalized learning (Lin, Yeh, Hung, & Chang, 2013). Having a learning path to be appropriate for all learners is impossible (Al-Muhaideb & Menai, 2011; Chen, 2009), and the inappropriate curriculum sequencing leads to cognitive overload even disorientation (Chen, 2008). Such issue has attracted an increased research interest, for examples, course generator is to assemble the courses sequence depending on learner's competence and learning goal (Ullrich & Melis, 2010), and learning path is to provide learning contents with pedagogical requirements meeting learner's profile and preference (Al-Muhaideb & Menai, 2011). Moreover, Garrido and Onaindia (2013) have indicated the challenge is to select the proper learning objects, define their relationships, and assemble their sequencing according to the learning goal and learner's status.

In the past, many researches have been done on customized learning regarding course composition and learning path considering learning goal, material difficulty, concept continuity and balance, limited time, learning ability, learner's profile, preference, need, test result, knowledge, cognitive style, learning style, and so on. Chen (2008) have utilized genetic algorithm and later Chen (2009) have combined it with ontology-based concept map. Chen and Duh (2008) have proposed the fuzzy item response theory to concern uncertain responses. Kontopoulos, Vrakas, Kokkoras, Bassiliades, and Vlahavas (2008) have used artificial intelligence and semantic web technology. Ullrich and Melis (2009) have presented a framework with pedagogical knowledge and later Ullrich and Melis (2010) have constructed a course generator with six different scenarios. Wong and Looi (2009) have presented the rule-based prescriptive planning and ant colony optimization-based inductive planning. Wang and Tsai (2009) have proposed a greed-like materials sequencing approach with discrete particle swarm optimization. Wang, Tseng, and Liao (2009) have employed the decision tree algorithm.

Moreover, Carchiolo, Longheu, and Malgeri (2010) have proposed a model to search for reliable resources suggested by peers. Chu, Chang, and Tsai (2011) and Li, Chang, Chu, and Tsai (2012) have adopted genetic algorithm and particle swam optimization. Klasnja-Milicevic et al. (2011) have applied aprioriall algorithm. Jeong, Choi, and Song (2012) have adapted the decision support system. Tan et al. (2012) have employed genetic algorithm with a layered topological sort algorithm. Chang and Ke (2013) have proposed a forcing legality operation in genetic algorithm to increase search efficiency. Durand, Belacel, and LaPlante (2013) have utilized the graph theory. Garrido and Onaindia (2013) have used planning techniques satisfying the temporal and resource constraints. Lin et al. (2013) have utilized hybrid decision trees. Furthermore, in the emerging u-learning learners interact real-world learning objects by context-aware technology (Hwang, Kuo, Yin, & Chuang, 2010b). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.