MASCE: A multi-agent system for collaborative e-learning

Abstract

E-learning has become one of the most popular teaching methods in recent years. One of its modes is the blended learning where learners can read teaching materials asynchronously from a teaching website and collaborate with their peers, while providing for necessary face-to-face explanation, discussion, and physical operation in the classroom. In the computational intelligence field, the Intelligent Agent paradigm gained a tremendous interest in many application domains over the last two decades [1, 2]. This research project paper focuses on the use of intelligent agents in the sphere of e-learning education with the help of collaborative learning [3]. Intelligent agents - the so called e-assistants or helper programs - can sit inside a computer and make the learning in e-learning happen dynamically to suit the need of the user. They can trap the user's likes and dislikes in various areas, the level of knowledge and the learning style and accordingly recommend the best matching helpers for collaboration. The paper introduces a Multi-Agent System for Collaborative E-learning (MASCE). MASCE is to assist teaching and learning process and also to encourage collaborative learning among peers. This system shall be used in a blended learning environment as a supplement to the face-to-face lecture where students can use the system in the lab or from home after attending the traditional lecture in the faculty. The objective is to incorporate the intelligence of the multi-agent system (MAS) in a way that enables it to actively and intelligently support the educational processes, where multiple agents can interact to exchange information so that students may collaborate on how best to gain knowledge. The proposed MAS system considers two types of users; namely students and instructors. Each of these users has a corresponding agent. These are Student Agent and Instructor Agent. The Student Agent runs on the student's computer. Thus, each student is equipped with a Student Agent, which helps the learning process of the student. It manages the student's personal profile and also tracks the student actions during learning process and updates his profile accordingly. On the other hand, the Instructor Agent provides teaching materials, assesses the progress and participation of different students through quizzes, and manages the progress of the course. The innovation in the proposed system is the introduction of the Assistant Agent which is initialized as soon as any of the users starts to use the system. The Assistant Agent runs on the system's server. It plays a centric role in the proposed system. The Assistant Agent has a collaboration mechanism which will be used for "match-making" and "community-building" to help increase collaboration between peers. It shall also give hints to the instructor to help in the teaching process such as statistics of the results of quizzes and summaries of students' profiles to help in the final grading. It acts a mediator between Student Agents and Instructor Agent. After receiving the preferences (goals) of the instructor and the students, it will run autonomously and self-dependently. Thus the proposed MAS consists of three types of agents. The course material is going to be structured in a hierarchical form where the course is divided into chapters and each chapter is divided into sections which in turn are divided into subsections and so on until we reach the leaves (concepts which cannot be divided any further). For each of these leaves the following will be provided: 1. Teaching materials 2. Quizzes to test student's knowledge level 3. Students' notes (blogs) 4. Discussion Forums 5. Questions asked by students requiring help The student can review all the teaching materials provided and add notes to his blogs if he wants to. He will take quizzes after each module to test his understanding (knowledge level) in that part so as to update his profile. If the student asks a question in a particular section, the Assistant Agent (Match Maker) will try to find the best potential helper for that question who is currently available online, willing and able to provide help. The Assistant Agent uses the students' models in that match making process. Some of the parameters which are going to be used to in the model are static, they are collected from the student himself through a questionnaire given to the student when he first uses the system including: 1. Help willingness 2. Initial availability 3. Preferences such as cognitive style Maximum numbers of concurrent discussions, 4. Initial belief of the student knowledge level through a simple quiz given to the student to classify him as either: novice, beginner, intermediate or advanced. 5. Weighted importance of various attributes: such as if he requires help quickly from any available willing helper, or he would rather wait to be matched with the helper with the best knowledge level in the concept he is asking about. Some of these parameters are dynamic; they are updated dynamically as the student interacts with the system and more new information is collected. Old information may be outdated or even wrong. For example, after each help session between two students, an evaluation form is presented to each of them to evaluate his colleague. These peer evaluations along with the collected information about student by the tracking system such as rate of his responses, are all used to update the helpfulness parameter. The actions (parameters) that the tracking system will monitor and that will be collected, modeled and analyzed are: 1. Actual Availability 2. Frequency of logging to the system (number of times in one week period for example) 3. Banned topics or users 4. Preferred users 5. Quizzes taken to update student's knowledge level. 6. Downloaded teaching materials 7. Blogs visited and notes added by the student to his blog 8. Number of postings on the forums 9. Frequency and type of questions asked to instructor or peers (not content-based) 10. Number of help requests accepted or rejected 11. Peer evaluation For finding best potential helper, the match maker (Assistant Agent in our system) first searches for those users online and who have not reached the maximum of their concurrent discussions online, and who are helpful and have good knowledge level of the concept asked about. The Multi-Agent System for Collaborative E-learning (MASCE) shall be installed on the lab that is equipped with 25 state-of-the-art PCs. The following servers are used to allow collaboration: Microsoft SharePoint Server, Microsoft Exchange Server, Microsoft Live Communication Server, Active Directory Server and Microsoft SQL Server. The lab is connected as an intranet; the access to the internet extends the benefits of the system outside the lab. The complete MAS system will provide functionalities essential in the educational process, such as real-time, as well as offline data and information gathering, analysis and distribution, embedded feedback, assessment, and collaboration. The complete system analysis is going to be the main part of a Ph.D. work. The first phase of this complete system will consider mainly the development of the basic functionality of the Student and Instructor Agents; and focus on the building and complete functionality of the Assistant Agent. The analysis and design phase of MASCE is done using Beliefs, Desires, Intentions-Agent Based Software Development (BDI-ASPD). We find desires first from the system requirements and then find its intention and corresponding belief. This idea comes from the natural approach we usually do in the real world. An agent's beliefs are a set of data describing the state of the environment. They are the knowledge that intentions use to fulfill their goals (desires). © 2008 IEEE.