Ontology-based curriculum content sequencing system with semantic rules

Abstract

Curriculum content sequencing involves managing a learning route to help users achieve learning goals. A conventional learning route consists of ordered content and is primarily based on a single course material. In an e-learning system, amount of similar course contents are available. These contents are expected to mutually substitute for one another in creating flexible learning routes. Owing to inconsistency in materials editing and cataloging, composing contents based on multiple sources leads to sequencing complexity. Most significantly, most e-learning systems lack a sequencing mechanism for dominating content composition. This study utilizes a knowledge-intensive approach to create a general sequencing knowledge base. This approach includes two components: (1) ontology is used to represent abstract views of content sequencing and course materials and (2) added semantic rules are used to represent relationships between individuals. Following knowledge base creation, both practical curriculum sequences and course materials can be inserted as factual knowledge. A reliable knowledge base can be established using inference power. An example involving mathematics course in elementary school education is designed using Web Ontology Language (OWL) and Semantic Web Rule Language (SWRL). Experimental lessons demonstrate that semantic rules in conjunction with ontologies not only solve sequencing problems but also achieve a durable knowledge base and a reliable system.

Introduction

Electronic learning (e-learning) is an instructional technology that relies on the Internet as the communication and presentation media. With the progress of the Internet, e-learning is rapidly growing in distance learning applications, which do not require the physical presence of learners. Instruction can be either synchronous or asynchronous. Researches have shown that e-learning provides effective and convenient solutions that facilitate sharing, integration and reuse of courseware (Alexander, 2001, Cloete, 2001, Lin and Hsieh, 2001). Some advanced challenges, for example the requirements of data exchange standard and courseware integration, emerge from the fundamental success of e-learning. The Shareable Content Object Reference Model (SCORM) is an example of such a standard and makes numerous digital learning objects available online. e-Learning is a current trend in providing enhanced services capable of outperforming traditional classroom teaching (Kaltenbach and Guo, 2001, Zhang et al., 2004). e-Learning brings changes not only in digitalized materials but also in learning styles and pedagogical activities. However, current e-learning technologies primarily focus on courseware building rather than learning services such as content sequencing. Consequently, in-depth content sequencing issues related to e-learning are worth discussing.

The curriculum content sequencing (learning route or route finding) problem is simple to understand, but complex enough to implement while the learning objects are given by a wide range of course providers. In practice, providing users with a suitable learning guidance that incorporates a clear and effective sequence for achieving their goal is important. In a face-to-face classroom, an instructor teaches a course simply using a textbook and a syllabus that covers the course in sequence. Instructors usually design a conventional learning route. Students then follow a fixed list of study contents. Since students have minimal alternatives among which they can choose, individual interests and preference are generally ignored (Chen, Liu, & Chang, 2006). Conversely, the growth of Internet usage worldwide lets learning more convenient. Learning objects with the same curriculum goal increase the selectiveness of sequence composition. However, owing to the inconsistency in materials editing and cataloging, composing these objects from multiple publishers results in sequencing complexity. Since most e-learning systems lack high level sequencing knowledge and relevant mechanisms, a content sequence is generally determined by the publishers rather than the system. A curriculum may comprise various learning sequences each with distinct publishers. Consequently, a flexible learning sequence containing substituted learning objects is a challenge.

A curriculum content sequence outlines the specific learning objects required along with the order in which they must be taken. Before composing a sequence based on objectives from multiple publishers, widely-accepted sequencing knowledge should be identified to better guide the objects. Since the content sequencing needs a high level guidance, knowledge-intensive approaches such as ontology are suited to this situation. Ontology is an emerging technology for implementing semantic knowledge that has been applied in expert systems (Guarino, 1997). Ontologies are typically utilized to establish taxonomy of a task domain and mold it into a conceptual structure (Chi, 2007). In this study, for example, elementary school mathematics courses are taken as the task domain. Meanwhile, the conceptual structure is abstract views of content sequencing activities, which can be used to internally formalize practical learning objects and externally express the relationships of learning objects at the knowledge level (Chandrasekaran, Josephson, & Benjamins, 1999). This study employs Web Ontology Language (OWL) as the notation for representing content sequencing knowledge. Since current OWL has only a limited ability to represent relationships between individuals, for example in role chaining, the rules are used to complement the inference of OWL ontology. A missing rule layer can be added above a solid foundation of the Semantic Web layer cake developed by Berners-Lee, Hendler, and Lassila (2001). Semantic Web Rule Language (SWRL) was recently proposed to increase the power of OWL. SWRL rules provide procedural knowledge to reduce the limitations of ontology inference, particularly in identifying semantic relationships between individuals (Horrocks, Patel-Schneider, Bechhofer, & Tsarkov, 2005). This study utilizes the combination of OWL and SWRL to address difficulties related to content sequencing. A walkthrough example is given to present how ontology and semantic rules can help build a content sequencing in e-learning system.

The remainder of this paper is organized as follows: Section 2 describes issues in curriculum planning and content sequencing. Section 3 then outlines the knowledge-intensive design using OWL ontology and SWRL rules. Subsequently, Section 4 provides implementation details related to knowledge base building. Section 5 then describes knowledge maintenance and knowledge retrieval by developing application interfaces. Next, Section 6 presents the discussion and conclusions of this study.