Academic journal article Educational Technology & Society

Exploring Learner Attitudes toward Web-Based Recommendation Learning Service System for Interdisciplinary Applications

Academic journal article Educational Technology & Society

Exploring Learner Attitudes toward Web-Based Recommendation Learning Service System for Interdisciplinary Applications

Article excerpt

Introduction

The huge quantity of learning resources currently available online has led to information overload, and traditional search engines can no longer meet the needs of all e-learners (Howe & Dreilinger, 1997; Pinkerton, 2000; Yan & Garcia-Molina, 1995; Zamboni, 1998) because these tools often provide irrelevant information while ignoring related content (Goldberg, Nichols, Oki, & Terry, 1992). These so-called recommendation systems aim to identify the appropriate learning materials among the e-learning resources available to help learners make appropriate choices (Resnick & Varian, 1997). Although only high-quality materials are presumably recommended to learners, such materials may not meet learners' expectations. For this reason, most current studies have focused on identifying information that meets the particular needs of learners (Wei, Moreau, & Jennings, 2005).

Current educational recommendation services use either explicit or implicit ratings. Calculations of explicit ratings are based on learners' input on such issues as interest in courses, quality of learning units, and difficulty of teaching materials. Thus, although the accuracy of these recommendations is high, excess data may present a burden to learners. Lang (1995) proposed a recommendation service that rated articles via browsers that would analyze and reorganize the ratings of previous readers to offer further recommendations to learners who have not yet read the articles (Krulwich & Burkey, 1997). Unlike explicit ratings, which are based on learners' expressed learning preferences, implicit ratings automatically record learning paths (e.g., learning materials read by learners, time spent on each learning unit, and frequency of daily visits to the learning website) for analysis. However, greater numbers of paths would require longer amounts of time for the calculation of recommendation rules; such implicit ratings are usually used to develop personalized recommendations for websites. A typical example of an implicit rating system was presented by Rucker and Polanco (1997), who suggested that the addition of a website to a list of bookmarks indicates interest in that website. Therefore, similar preferences can be calculated on the basis of the bookmarked sites and used to recommend additional websites.

Previous research on the recommendation of learning services has usually focused on recommendations for one course, with the goal of continuously creating different recommendation algorithms to enable learners to obtain more precise content (Johannes, Matthias, Christoph, & Ralf, 2008; Tasi, Chiu, Lee, & Wang, 2006; Wei et al., 2005). However, these studies have neglected immediate individual interactive learning models based on interdisciplinary learning, new fields of learning, and changes in academic interests. Content restricted to single subject cannot meet the diverse needs of modern industrial development, and interdisciplinary learning and cooperation have emerged as approaches for learners trying to solve complicated professional issues. From a psychological perspective, interaction between different fields and domains of knowledge can trigger learners' intellectual potential, and exposure to multiple domains can enhance the scope, depth, and novelty of knowledge (Johansson, 2006). Thus, this study examined a platform for a recommendation service that supports interdisciplinary learning and used the advantages of explicit and implicit ratings to develop a mechanism for detecting changes in academic interests. Such a mechanism enabled the system to recognize each learner's learning process and to actively recommend content from the specific fields required by learners. It also provided precise recommendations for learning content and reused pedagogical resources.

This study was designed to implement an interdisciplinary recommendation system that provides educational suggestions while monitoring changing academic interests to meet the individualized interactive learning demands of learners by recommending content that is appropriate to their current needs. …

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.