CNN in Practice: A Guide to Evaluating Multiple-Choice Exams
Keywords:
Convolutional Neural Networks, Artificial Intelligence, Automated Evaluation, Image Processing, Deep Learning, Public SectorAbstract
This guide presents a convolutional neural network (CNN)-based system for automating the evaluation of multiple-choice exams. Python and TensorFlow were used to develop a model capable of detecting answers with high accuracy (99.18%), overcoming the limitations of digital image processing (DIP) and a recall of 99.45%. The results show the model's effectiveness in automatic answer classification. The research highlights the importance of image preprocessing and the selection of the activation function based on the problem of enhancing performance. Although the system is efficient, it faces limitations related to dataset diversity and the need for specialized hardware. Future improvements include expanding the dataset and optimizing the model for low-resource devices. This work contributes to the field of artificial intelligence applied to education by facilitating automated exam evaluation and optimizing grading time.References
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., ... & Zheng, X. (2016). TensorFlow: A system for large-scale machine learning. 12th {USENIX} symposium on operating systems design and implementation ({OSDI} 16), 265-283.
Bishop, C. M. (2006). Pattern recognition and machine learning. Springer.
Chen, J., Qi, X., Wang, W., Li, B., & Liu, Y. (2020). Real-time location of surgical incisions in cataract phacoemulsification. Springer Science+Business Media, LLC, part of Springer Nature. https://doi.org/10.1007/s0010-2020
Chollet, F. (2017). Deep learning with Python. Manning Publications Co.
Cordón Luis, S. (2019). Reconocimiento de torres de alta tensión mediante algoritmos de visión por computador e inteligencia artificial: LR, SVM, CNN.
De Carvalho, A. S. A., Pereira, E. C. D. S., Da Silva, E. J., & Piva Jr, D. (2022). Análise comparativa entre os principais algoritmos de detecção facial: Haar Cascade, HOG, CNN, YOLO e DeepFace. In OPEN SCIENCE RESEARCH V (Vol. 5, pp. 439-454). Editora Científica Digital.
Glorot, X., & Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks. Proceedings of the thirteenth international conference on artificial intelligence and statistics, 249-256.
Gómez Pujante, B. (2023). Redes convolucionales. Aplicación a la clasificación de imágenes médicas [Trabajo de fin de grado, Universidad Miguel Hernández de Elche]. Universidad Miguel Hernández de Elche. https://hdl.handle.net/11000/30233
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.
Haykin, S. (2009). Neural networks and learning machines (Vol. 3). Pearson Education.
Heaton, J., Goodfellow, I., Bengio, Y., & Courville, A. (2018). Deep learning. Genetic Programming and Evolvable Machines, 19(3-4), 305–307. https://doi.org/10.1007/s10710-017-9314-z
Kanabur, V., Harakannanavar, S. S., Purnikmath, V. I., Hullole, P., & Torse, D. (2020). Detection of leaf disease using hybrid feature extraction techniques and CNN classifier. In Computational Vision and Bio-Inspired Computing: ICCVBIC 2019 (pp. 1213-1220). Springer International Publishing.
Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. *arX
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. In F. Pereira, C. J. Burges, L. Bottou, & K. Q.
Weinberger (Eds.), Advances in Neural Information Processing Systems (Vol. 25). Curran Associates, Inc. https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf
LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015).
Maas, A.L., Hannun, A.Y. and Ng, A.Y. (2013) Rectifier Nonlinearities Improve Neural Network Acoustic Models. Proceedings of the 30th International Conference on Machine Learning, Vol. 28, 3. https://ai.stanford.edu/~amaas/papers/relu_hybrid_icml2013_final.pdf
Nair, V., & Hinton, G. E. (2010). Rectified linear units improve restricted boltzmann machines. Proceedings of the 27th international conference on machine learning (ICML-10), 807-814.
Nama, P. (2022). Optimizing automation systems with AI: A study on enhancing workflow efficiency through intelligent decision-making algorithms. World Journal of Advanced Engineering Technology and Sciences, 7(02), 296-307.
Pardo, Á. (s.f.). Filtrado de imágenes. Departamento de Ingeniería Eléctrica, Universidad Católica del Uruguay. Recuperado de https://grupivis.dc.uba.ar/archivos/UBA_FiltradoImagenes.pdf
Peña, L. J. M. (2016). HABCO: Herramienta informática para la automatización de la calificación de exámenes para apoyo al docente. Res. Comput. Sci., 111, 9-21.
Ramos Armijos , D. F., Ramos Armijos , D. G., Ramos Armijos , N. J., Tapia Puga , V. M., & Tapia Puga , L. I. (2024). Explorando las Fronteras: la Aplicación de Inteligencia Artificial en la Evaluación Educativa. Ciencia Latina Revista Científica Multidisciplinar, 7(6), 5657-5672. https://doi.org/10.37811/cl_rcm.v7i6.9108
Russell, S. J., & Norvig, P. (2004). Inteligencia artificial: Un enfoque moderno (2ª ed.). Pearson Educación.
Shorten, C., Khoshgoftaar, T.M. A survey on Image Data Augmentation for Deep Learning. J Big Data 6, 60 (2019). https://doi.org/10.1186/s40537-019-0197-0
Van Rossum, G., & Drake, F. L. (2009). Introduction to python 3: python documentation manual part 1. CreateSpace.
