A New Paradigm for Intelligent Tutoring Systems: Example-Tracing Tutors

By Vincent A. W. M. M. Aleven, Bruce M. McLaren, Jonathan Sewall, K. Koedinger. International Journal of Artificial Intelligence in Education, 2009

348 citations recorded by the source corpus.

Key success criteria for an ITS authoring tool are that (1) the tool supports the creation of effective tutoring systems, (2) the tool can be used to build tutors across a wide range of application domains, (3) authoring with the tool is cost-effective, (4) the tool supports easy deployment and delivery of tutors in a variety of technical contexts, (5) tutors created with the tool are maintainable, and (6) if tutors are used in a research context, the tool must support research-related functionality. The Cognitive Tutor Authoring Tools (CTAT) address all of these requirements to a substantial degree, fully meeting most of them. CTAT supports the creation of both Cognitive Tutors (Koedinger & Corbett, 2006) and a newer type of tutors called example-tracing tutors. This paper focuses on the latter. Example-tracing tutors evaluate student behavior by flexibly comparing it against examples of correct and incorrect problem-solving behaviors. Example-tracing tutors are capable of sophisticated tutoring behaviors: they provide step-by-step guidance on complex problems while recognizing multiple student strategies and maintaining multiple interpretations of student behavior. On that basis, they should be deemed intelligent tutoring systems. Example-tracing tutors can be built without programming, through drag-and-drop techniques and programming by demonstration. Exampletracing tutors have been built and used in real educational settings for a wide range of application areas. Development time estimates from a large number of projects that have used CTAT suggest that CTAT improves the cost-effectiveness of ITS development by a factor of 4-8, compared to “historical” estimates of tutor development time. Although there is a lot of variability in these kinds of estimates, they nonetheless support our hope that lowering the skill requirements for tutor creation is a key step toward widespread use of ITS technology. The main contributions of the work are the example-tracing tutor technology and tools for building these types of tutors without programming.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/ff0c22178fd0e8602af4dd698a150fcba5e39970

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

AutoTutor and Family: A Review of 17 Years of Natural Language Tutoring (2014, background): “Other successful tutoring systems, such as Cognitive Tutor (Aleven et al. 2009; Ritter et al. 2007) and Andes (VanLehn et al.”
New Potentials for Data-Driven Intelligent Tutoring System Development and Optimization (2013, methodology): “In this approach, authors first use Cognitive Tutor Authoring Tools (CTAT; Aleven et al. 2009) to create a graphical user interface which students will use to solve tasks (e.g., a table of rows for steps in an algebra equation solution).”
Code comprehension problems as learning events (2012, methodology): “The feedback was generated automatically by the system using a platform designed with the Cognitive Tutor Authoring Tools (CTAT) [20].”
A Systematic Literature Review of Automated Feedback Generation for Programming Exercises (2018, background): “Example-tracing tutors follow the steps that a student takes based on “generalised examples of problem-solving behaviour” [4], making it easier to create ITSs.”