Algorithmic strategy to assess, cluster, and predict university physics learning profiles with educational robotics
Main Article Content
Abstract
Introduction. Educational robotics enables the exploration of physical principles through programmable platforms; however, individual differences in interaction can be hidden by group averages. Objective. An algorithmic strategy was designed to assess conceptual learning of Newtonian mechanics, identify interaction profiles, and predict gain categories using an omnidirectional robotic platform. Methodology. A methodological and algorithmic-proposal approach was established, based on: data preparation, calculation of individual normalized change and comparison between guides, k-means clustering with interaction variables, profile characterization, gain classification, and prediction with random forest. The procedure was verified using simulated data from a control group and an experimental group, with 30 students per group, in a reproducible Google Colab environment, and was evaluated with multiple simulated datasets through profile recovery, clustering stability, and model accuracy. Results. In the simulated scenario, the medians of normalized change were 0.55 in the experimental group and 0.25 in the control group. Three interaction profiles with differentiated gains (0.83, 0.59, and 0.14) were recovered, and the predictive model outperformed the majority class (0.76 versus 0.40). Clustering was stable in 98% of the 50 simulated datasets. Conclusions. The strategy made it possible to examine interaction patterns without attributing learning styles to students. Its results correspond to a demonstration with simulated data; its educational usefulness must be verified with real students and a validated assessment instrument.
General area of study: Physical Sciences. Specific area of study: Machine learning and data science applied to physics teaching. Type of study: Original article.
Downloads
Article Details
References
Aldowah, H., Al-Samarraie, H., y Fauzy, W. M. (2019). Educational data mining and learning analytics for 21st century higher education: a review and synthesis. Telematics and Informatics, 37, 13–49. https://doi.org/10.1016/j.tele.2019.01.007
Arvin, F., Espinosa, J., Bird, B., West, A., Watson, S., & Lennox, B. (2019). Mona: an affordable open-source mobile robot for education and research. Journal of Intelligent & Robotic Systems, 94, 761–775. https://doi.org/10.1007/s10846-018-0866-9
Boya-Lara, C., Saavedra, D., Fehrenbach, A., & Márquez-Araque, A. (2022). Development of a course based on BEAM robots to enhance STEM learning in electrical, electronic, and mechanical domains. International Journal of Educational Technology in Higher Education, 19, Article 7. https://doi.org/10.1186/s41239-021-00311-9
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
Eliza, F., Hakiki, M., Muhtaj, M., Putri, D. A., Hidayah, Y., Fricticarani, A., Fakhri, J., Arpannudin, I., Subroto, D. E., Sussolaikah, K., Hamid, M. A., Fadli, R., & Ramadhan, M. A. (2025). Game-D: development of an educational game using a line follower robot on straight motion material. International Journal of Information and Education Technology, 15(1). https://doi.org/10.18178/ijiet.2025.15.1.2217
Hake, R. R. (1998). Interactive-engagement versus traditional methods: a six-thousand-student survey of mechanics test data for introductory physics courses. American Journal of Physics, 66(1), 64–74. https://doi.org/10.1119/1.18809
Han, J., Bao, L., Chen, L., Cai, T., Pi, Y., Zhou, S., Tu, Y., y Koenig, K. (2015). Dividing the force concept inventory into two equivalent half-length tests. Physical Review Special Topics - Physics Education Research, 11(1), 010112. https://doi.org/10.1103/PhysRevSTPER.11.010112
Hubert, L., & Arabie, P. (1985). Comparing partitions. Journal of Classification, 2(1), 193–218. https://doi.org/10.1007/BF01908075
López-Neri, E., Luque-Vega, L. F., González-Jiménez, L. E., & Guerrero-Osuna, H. A. (2023). Design and implementation of a robotic arm for a MoCap system within extended educational mechatronics framework. Machines, 11(9), Article 893. https://doi.org/10.3390/machines11090893
Manu, F. A., Ramaila, S., & Ramnarain, U. (2025). The role of Arduino-assisted robotics coding application in physics learning: a systematic review. International Journal of Science, Mathematics and Technology Learning, 32(2), 189–211. https://doi.org/10.18848/2327-7971/CGP/v32i02/189-211
Martínez-Prado, M. A., Rodríguez-Reséndiz, J., Gómez-Loenzo, R. A., Camarillo-Gómez, K. A., & Herrera-Ruiz, G. (2019). Towards a new tendency in embedded systems in mechatronics for the engineering curricula. Computer Applications in Engineering Education, 27, 603–614. https://doi.org/10.1002/cae.22101
Marx, J., & Cummings, K. (2007). Normalized change. American Journal of Physics, 75(1), 87–91. https://doi.org/10.1119/1.2372468
Nasriddinov, D. K., Alimova, S., Chorieva, M. J., Kenjayeva, X., y Ochilova, F. B. (2026). Pedagogical experimental work on the effectiveness of teaching physics based on mobile applications and robotics elements. Natural Resources for Human Health, 6(5S), 353–361. https://doi.org/10.53365/nrfhh.577
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830. https://jmlr.org/papers/v12/pedregosa11a.html
Rousseeuw, P. J. (1987). Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53–65. https://doi.org/10.1016/0377-0427(87)90125-7
Růžicka, M., Rejthárek, M., & Major, Š. (2018). Project-based learning in basic course of technical physics: Computer-controlled experiments and Agros2D modeling. WSEAS Transactions on Applied and Theoretical Mechanics, 13, 46–52. https://wseas.com/journals/articles.php?id=2475
Scaradozzi, D., Cesaretti, L., Screpanti, L., y Mangina, E. (2020). Identification of the students learning process during education robotics activities. Frontiers in Robotics and AI, 7, Article 21. https://doi.org/10.3389/frobt.2020.00021
Shariati, A., Shamekhi, A. H., Ghaffari, A., Gholampour, S., & Motaghed, A. (2019). Conceptual design algorithm of a two-wheeled inverted pendulum mobile robot for educational purposes. Mechanics of Solids, 54, 614–621. https://doi.org/10.3103/S0025654419040162
Verner, I. M., Cuperman, D., Gamer, S., & Polishuk, A. (2020). Exploring affordances of robot manipulators in an introductory engineering course. International Journal of Engineering Education, 36(5), 1691–1707. https://www.ijee.ie/1atestissues/Vol36-5/22_ijee3981.pdf
Yang, L., Xiao, S., Wang, T., Shi, Y., y Driggs-Campbell, K. (2024). Robotics via experiential active learning with immersive technology: a pedagogical framework designed for engineering curricula. IEEE Access, 12, 65706–65715. https://doi.org/10.1109/ACCESS.2024.3393473
Zhou, D., Xie, M., Xuan, P., y Jia, R. (2020). A teaching method for the theory and application of robot kinematics based on MATLAB and V-REP. Computer Applications in Engineering Education, 28, 239–253. https://doi.org/10.1002/cae.22188.