Detecting Learners’ Weak Points Utilizing Time Intervals of Pen Strokes

  • Kazuya Kishi Kyushu Institute of Technology
  • Motoki Miura Kyushu Institute of Technology
Keywords: digital pen, weak point, writing data


We consider that most learners tend to focus on and review only problems they incorrectly answered after performing exercises. To deepen their understanding, it is also necessary for learners to review problems that took time to answer even though they were answered correctly. However, it is difficult to judge which problems took time to answer with an ordinary pen and paper. Therefore, we adopt a digital pen to help learners to recognize parts of a problem that took long to complete. Using the pen-stroke interval data obtained by a digital pen, we can discover the weak points of a learner. In this study, we implemented the method and evaluated it by comparing the weak points extracted by system, learners and evaluators. The results confirm that the method can detect weak points with an accuracy of about 50% to 60%.


Ministry of Education, Culture, Sports, Science and Technology. The Vision for ICT in Education - Toward the Creation of a Learning System and Schools Suitable for the 21st Century -., April 2011.

Roundtable on informatization of education for the 2020s. “Roundtable on informatization of education for the 2020s” final summary., 2016.(in Japanese).

Motoki Miura, Taro Sugihara, and Susumu Kunifuji. Improvement of digital pen learning system for daily use in classrooms. In Educational Technology Research, Vol. 34, pp. 49-57, 2011.

Seiichi Uchida, Marcus Liwicki, Masakazu Iwamura, Koichi Kise, and Shinichiro Omachi. Digital pen. In Journal of the Institute of Image Information and Television Engineers,Vol.64,No.3, pp. 293-298, 2010.(in Japanese).

Hiroki Asai, Akari Nozawa, Shogo Sonoda, and Hayato Yamana. Student’s stumble detection using online handwriting data. In DEIM Forum A8-4, 2012.(in Japanese).

Hiroki Asai and Hayato Yamana. Detecting student frustration based on handwriting behavior. UIST’ 13.

Kun Yu, Julien Epps, and Fang Chen. Mental workload classication via online writing features. In 12th International Conference on Document Analysis and Recognition, vol. 00, pp. 1110-1114, 2013.

Chihiro Nakatsuka, Yoshitaka Morimura, and Atsushi Hashimoto. Detecting answer stagnation point using time intervals of pen strokes. In JSiSE research report 30(7), pp. 71-74, March 2016.(in Japanese).

Hiroyuki Deguchi, Yuya Onishi, Tetsuya Ohya, Hironori Koyama, and Masashi Kawasumi. Review support system which selects the questions by using answering time. In FIT2010, 2010.(in Japanese).