An adaptive and innovative question-driven competition-based intelligent tutoring system for learning

Abstract

ISCARE (Information System for Competition based on pRoblem solving in Education) is a new and innovative intelligent tutoring system that we have designed and implemented. This tool allows the competition among students for improving their learning process in a course. The tool takes some ideas from the Swiss-system widely used in chess and adapts them to the educational area. The competition is based on different tournaments and rounds. In each round, students are assigned in pairs of two, which compete one against another, and each pair receives different questions that students have to solve in a limit of time. Students can see their partial ratings after each round and their final rating after a tournament. A lot of knowledge from different disciplines was used to design, and implement this system, as ISCARE includes different functionality such as the students’ registration into the system, the creation of tournaments, the registration and assignment of students to tournaments, the management of each tournament life cycle (started, in execution, finished, etc.), the addition of the different exercises to tournaments, the calculation of pairs of students for each round with different algorithms, the assignment of exercises per round and pair, the scorings of the students per round and tournament, the management of the students’ ratings, or the visualization of information. This paper presents the ISCARE intelligent tutoring system, describing its different options, menus, or functionality as well as its architecture and the specific modeling to achieve the desired features.

Introduction

The use of intelligent tutoring systems (ITSs) in education has increased considerably over the last years. An ITS is an expert system that helps students and teachers in the learning process. ITSs can be accessed by teachers and students through the Web or as a stand-alone application. An ITS is usually divided into four modules (Wenger, 1987): the interface module (what teachers and students interacts with in the system), the expert module (the specific contents), the pedagogical module (the different strategies of teaching), and the user model (some data to represent students and teachers in the system, so that different contents and strategies can be adapted to the different profiles). ITSs are software applications that must enable and manage the desired learning processes, and make convenient decisions about different issues such as the materials that are shown at each moment for different profiles, or motivational and emotional aspects.

One type of ITS is based on the competition among students in their learning process, but there are only a few of these competition-based ITSs. Maybe, this limited number of competition-based ITSs is a consequence of the different arguments against competition, which are described by different authors, e.g. (Kohn, 1986). But there are other studies that argue about the benefits of competition in education, e.g. (Yu, 2001). From our point of view, the success of the introduction of competitive tools in the learning process depends on the specific competitive strategies applied, as stated in Ediger (1996).

In this direction, we have developed and implemented a system for students’ competition in education that tries to include some of the best practices to increase student learning in the process. The designed and implemented competition-based system is called ISCARE (Information System for Competition based on pRoblem solving in Education), and it includes a couple of new and innovative features with respect to existing competition-based systems for learning, as it is analyzed in Section 2.

One of the first ideas taken for the design of ISCARE was the application of the Swiss-system used e.g. in chess tournaments. The Swiss-system divides a tournament in different rounds. In each round, participants are paired, forming matches of one against one. Students are paired with classmates of their similar knowledge level in that moment. Participants can win (1 point), lose (0 points), or draw (0.5 points) in each round. All participants play the same number of games (one per round) even if they lose in several games. There are partial ratings after each round in which students are rated according to their performance up to that round, and a final tournament rating. The Swiss-system has been adapted in ISCARE taken into account an educational competition perspective. ISCARE supports different rounds in which students participate, and matches are assigned according to the students scorings up to that round, so that students with a similar knowledge level can compete one against another. In addition, students cannot only receive points to increase their scoring depending on the comparison with their opponents, but also depending on their problem solving performance in that round.

ISCARE has to take care of a lot of tasks for implementing the desired competition that can produce a learning benefit for students. For example, ISCARE has to make the student pair assignments for each round (trying to pair students with a similar scoring up to that moment), select the different possible problems for each round, adapt the different problems to the different pairs (trying to assign problems with a difficulty level according to the students pairs knowledge levels), track the different students’ progress during the tournament and their statistics (such as topics in which a student failed more), show the different problems to the students in a synchronized way, show the opponents and own information during the round competition, manage a tournament life cycle (start, execution, finished, etc.), manage the different tournaments (creation, deletion, assignment of students to different tournaments, etc.), or enabling teachers to upload different types of problems for students before a tournament starts and making possible their reuse.

Several of the presented tasks in the ISCARE system cannot be done efficiently by human teachers, without the intervention of computers. Teachers would have to devote a lot of time to execute the algorithms to pair the different students, adapt the different problems for different pairs, give different pieces of papers for every student which are adaptive depending on previous round results, make the different scoring calculations, manage the different rounds and tournaments, etc. In addition, these are complex operations that can usually lead to teachers’ mistakes because of the high number of algorithmic operations to perform. Nevertheless, an intelligent system such as ISCARE can make it quickly, efficiently and without errors because of the high computing capacity of ITSs. In order to achieve it properly, a lot of considerations from different expertise areas must be taken into account, which includes: a useful interface for teachers and students, a complete and proper modeling for all the educational and management functionality, selection of the proper algorithms (for adaptation of problems, adaptation of pairs, scoring calculations, etc.), or the selection of the architecture.

This paper shows a detailed description of the implemented ISCARE system to show all their features and the differences with respect to existing systems. Furthermore, the paper shows how the functionality is implemented as a specific modeling. Section 2 explains related works and compares existing competition-based ITSs for education with respect to our solution, emphasizing the novelties of our ISCARE system. Section 3 explains the architecture of ISCARE. Sections 4 Management of users and tournaments, 5 Description of a competition round in the system, 6 Information about the users and problems in the system are devoted to the explanation and description of the different features of ISCARE: Section 4 is about the management of users and tournaments in the system, Section 5 is about the description of a specific round, while Section 6 is about the different ratings and information of the users in the system. Finally, Section 7 summarizes our conclusions and directions of future work regarding the ISCARE tool.