Assistant for math reviewing
DOI:
https://doi.org/10.61173/5188n722Keywords:
math reviewing, computer-assisted learning, intelligent tutoring systemsAbstract
If students want to practice or review any knowledge, they may look for related exercises to improve their proficiency in the knowledge. But choosing exercises that fit students’ present proficiency by students themselves is difficult and wastes much time. The educational app: Assistant for math reviewing can find appropriate exercises automatically and dynamically for students to help them improve quickly and efficiently. This app gets students’ proficiency based on their performance in answering exercises and gives advice for students to review based on their proficiency. The author addresses the following questions: what kind of exercises or knowledge should students review? How often should students review? How Assistant for math reviewing helps students improve efficiently. This paper outlines how Assistant for math reviewing works, what benefits students can get from this app, why this app can offer benefits, and why this app is better than other tools for reviewing math.
References
[1] Mostow, J. (2012, June 6-8). Why and How Our Automated truly work? It is necessary to design experiments to evaluate the models and prevent their over-fitting because Reading Tutor Listens. In International Symposium on of the lack of data. The author uses a ten-fold cross- Automatic Detection of Errors in Pronunciation Training validation, dividing the students randomly into ten disjoint (ISADEPT), 43-52. KTH, Stockholm, Sweden. sets, as illustrated in Figure 2. App trains on nine sets, [2] Heiner, C., Beck, J. E., & Mostow, J. (2004, June 17-19). tests on the held-out set, repeats this procedure for each Improving the Help Selection Policy in a Reading Tutor that set, and averages the results(Heiner, 2004). We evaluate Listens. Proceedings of the InSTIL/ICALL Symposium on the model by comparing the average results with those NLP and Speech Technologies in Advanced Language Learning obtained from the models. For instance, If the model Systems, Venice, Italy, 195-198. calculates the best level of difficulty of questions that [3] Mostow, J. (2004, August 30). Some useful design tactics fits students. Each test set gets the best level of difficulty for mining ITS data. Proceedings of the ITS2004 Workshop L in training sets and the average growth rate gr of the on Analyzing Student-Tutor Interaction Logs to Improve proficiency in the test set by using L. Then we evaluate Educational Outcomes, Maceió, Alagoas, Brazil, 20-28. the model by comparing gr with the growth rate we get [4] Mostow, J. (2004, July 22). If I Have a Hammer: in the model. If these two results are similar, the model Computational Linguistics in a Reading Tutor that Listens will behave well in calculating the best level of difficulty [Invited keynote address]. 42nd Annual Meeting of the of questions to offer next. But if they are different, the Association of Computational Linguistics (ACL-EACL 2004), model may appear over-fitting, and we can choose a better Barcelona, Spain. parameter in the model by ten--cross-validation. [5] Mostow, S. Roth, A. G. Hauptmann, and M. Kane, “A Prototype Reading Coach that Listens,” Proceedings of the Twelfth National Conference on Artificial Intelligence (AAAI- 94), American Association for Artificial Intelligence, Seattle, WA, August 1994, pp. 785-792. Recipient of the AAAI-94 Outstanding Paper Award. Figure 2: Ten-fold cross-validation [6] Beck, J. E., Chang, K.-m., Mostow, J., & Corbett, A. (2008, June 23-27). Does help? Introducing the Bayesian Evaluation 6. Conclusion and Future Work and Assessment methodology. 9th International Conference on The author presents an educational app that assists Intelligent Tutoring Systems, Montreal, 383-394. ITS2008 Best students in reviewing math knowledge. This app, Assistant Paper Award.
Downloads
Published
Issue
Section
License
Copyright (c) 2023 by the authors.
This work is licensed under a Creative Commons Attribution 4.0 International License.
