VIDEO TUTORIAL PERSONALIZATION SYSTEM BASED ON MACHINE LEARNING
DOI:
https://doi.org/10.24867/06BE04DjericKeywords:
personalized teaching, eye-tracker device, clustering, KNN, KMeansAbstract
The problem that this paper solves is the recommendation of the most appropriate type of tutorial to the student. This paper is the result of research conducted on students of the Faculty of Technical Sciences. Twenty nine students were questioned. Students answered survey questions, watched video tutorials on sorting algorithms, and answered questions about sorting algorithms. When they were watching one of the video tutorials, their eye movement were followed using an eye-tracking device. In that way was formed dataset that was used to build the model. The model that gave the highest accuracy is consisted of KNN algorithm, which is applied in finding students with the most similar type of movement. The most appropriate type of tutorial is found based on similar students. This model achieved an accuracy 75% on the validation set and accuracy 67% on the test set.
References
[2] Mele, Maria Laura, and Stefano Federici. Gaze and eye-tracking solutions for psychological research. Cognitive processing, 2012, 13.1: 261-265.
[3] Sharafi, Zohreh, et al. Eye-tracking metrics in software engineering. In: 2015 Asia-Pacific Software Engineering Conference (APSEC). IEEE, 2015. p. 96-103.
[4] https://imotions.com/blog/7-terms-metrics-eye-tracking/ (pristupljeno u avgustu 2019.)
[5] Borys, Magdalena, and Małgorzata Plechawska-Wójcik. Eye-tracking metrics in perception and visual attention research. EJMT, 2017, 3: 11-23.
[6] Göbel, Fabian, and Henry Martin. Unsupervised Clustering of Eye Tracking Data. In: Spatial Big Data and Machine Learning in GIScience, Workshop at GIScience 2018. p. 25-28.
[7] Rousseeuw, Peter J. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. Journal of computational and applied mathematics, 1987, 20: 53-65.
[8] Maaten, Laurens van der; HINTON, Geoffrey. Visualizing data using t-SNE. Journal of machine learning research, 2008, 9.Nov: 2579-2605.
[9] Moss, Felix Joseph Mercer, Roland Baddeley, and Nishan Canagarajah. Eye movements to natural images as a function of sex and personality. PLoS One, 2012, 7.11: e47870.
[10] Astrachan, Owen. Bubble sort: an archaeological algorithmic analysis. In: ACM SIGCSE Bulletin. ACM, 2003. p. 1-5.
[11] Hayfron-Acquah, J. B., Obed Appiah, and K. Riverson. . Improved Selection Sort Algorithm. International Journal of Computer Applications, 2015, 110.5.
[12] https://www.gazept.com/product/gazepoint-gp3-eye-tracker/, (pristupljeno u avgustu 2019.)
[13] https://www.gazept.com/tutorials/, (pristupljeno u avgustu 2019.)
[14] https://www.python.org/, (pristupljeno u avgustu 2019.)
[15] https://towardsdatascience.com/choosing-the-right-encoding-method-label-vs-onehot-encoder-a4434493149b, (pristupljeno u avgustu 2019.)
[16] Wold, Svante, Kim Esbensen, and Paul Geladi. Principal component analysis. Chemometrics and intelligent laboratory systems, 1987, 2.1-3: 37-52
[17] Peterson, Leif E. K-nearest neighbor. Scholarpedia, 2009, 4.2: 1883.
[18] Krafka, Kyle, et al. Eye tracking for everyone. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. p. 2176-2184.
