"Many of the intelligent tutoring systems that have been developed during the last 20 years have proven to be quite successful, particularly in the domains of mathematics, science, and technology. They produce significant learning gains beyond classroom environments. They are capable of engaging most students' attention and interest for hours. We have been working on a new generation of intelligent tutoring systems that hold mixed-initiative conversational dialogues with the learner. The tutoring systems present challenging problems and questions to the learner, the learner types in answers in English, and there is a lengthy multiturn dialogue as complete solutions or answers evolve. This article presents the tutoring systems that we have been developing. AutoTutor is a conversational agent, with a talking head, that helps college students learn about computer literacy. andes, atlas, and why2 help adults learn about physics. Instead of being mere information-delivery systems, our systems help students actively construct knowledge through conversations."
Arthur C. Graesser, Kurt VanLehn, Carolyn P. Rosé, Pamela W. Jordan, and Derek Harter. 2001. Intelligent tutoring systems with conversational dialogue. AI Mag. 22, 4 (October 2001), 39-51.
1. SUMMARY
The paper describes AutoTutor which is a fully automated conversational-based tutoring system simulating typical human tutor’s dialogue patterns. From pre-designed curriculum scripts , AutoTutor gives questions that stimulate detailed answers with deep reasoning. A Dialog Advancer Network - a finite state automaton - was employed to drive conversations in turn-by- turn or multi-turn manner. It usually takes 10 to 30 turns to answer 1 question in the curriculum script. The process starts with a “pump” for information from a student. Based on the response, AutoTutor picks one piece of knowledge for further drilling, and sets expectation for student’s answer. Each expectation is usually satisfied within 6 dialog moves, which include hints, assertions, and prompts. Latent semantic analysis (LSA) was used to evaluate the quality of students’ responses. Feedbacks are given via backchannel feedback, evaluative pedagogical feedback, and corrective feedback.
2. STRENGTHS
The paper has a good scope with AutoTutor has helped 200 college students, with improved learning of 0.5 (half a letter grade). The paper also discusses ATLAS and WHY2 for comparing the pros and cons. Interesting findings were shared such as sometimes students can’t articulate a point even when such point was given away prior by the AutoTutor.
3. WEAKNESSES
Readers might desire for more quantitative data regarding AutoTutor. Intuitively, the learning rates will vary among different questions in the curriculum script. It would be interesting to see which question requires most students most time to answer, the percentage of cycles ended with assertions, and so on.
