Paper Information | SBC BOOKS

Fill in your paper information

Language (*)

Title (*)

Keywords (*)

Abstract (*)

The accuracy of machine learning models implementing classificationtasks is strongly dependent on the quality of the training dataset. This is a chal-lenge for domains where data is not abundant, such as personalized medicine,or unbalance, as in the case of images of plant species, where some species havevery few samples while others offer large number of samples. In both scenarios,the resulting models tend to offer poor performance. In this paper we presenttwo techniques to face this challenge. Firstly, we present a data augmentationmethod called SAGAD, based on conditional entropy. SAGAD can balance mi-nority classes in conjunction with the increase of the overall size of the trainingset. In our experiments, the application of SAGAD in small data problems withdifferent machine learning algorithms yielded significant improvement in per-formance. We additionally present an extension of SAGAD for iterative learningalgorithms, called DABEL, which generates new samples for each epoch usingan optimization approach that continuously improves the model’s performance.The adoption of SAGAD and DABEL consistently extends the training datasettowards improved target classification performance.

Pages (*)

File Link

English Information

Title

Keywords

Abstract

(*) To change the order drag the item to the new position.

Authors

#	Name
1	Henrique Silva(matheusf@lncc.br)
2	Rafael Silva(Rpereira@lncc.br)
3	Fabio Porto(Fporto@lncc.br)

(*) To change the order drag the item to the new position.

Reference

#	Reference
1	Ashrapov, I. (2020).Gans for tabular data.https://github.com/Diyago/GAN-for-tabular-data.
2	Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: syn-thetic minority over-sampling technique.Journal of artificial intelligence research,16:321–357
3	Chen, M., Hao, Y., Hwang, K., Wang, L., and Wang, L. (2017). Disease prediction bymachine learning over big data from healthcare communities.IEEE Access, 5:8869–8879.
4	Cugliari, G., Benevenuta, S., Guarrera, S., Sacerdote, C., Panico, S., Krogh, V., Tumino,R., Vineis, P., Fariselli, P., and Matullo, G. (2019). Improving the prediction of car-diovascular risk with machine-learning and dna methylation data. In2019 IEEE Con-ference on Computational Intelligence in Bioinformatics and Computational Biology(CIBCB), pages 1–4.
5	Dua, D. and Graff, C. (2017a). UCI machine learning repository.
6	Dua, D. and Graff, C. (2017b). UCI machine learning repository.
7	from Jed Wing, M. K. C., Weston, S., Williams, A., Keefer, C., Engelhardt, A., Cooper,T., Mayer, Z., Kenkel, B., the R Core Team, Benesty, M., Lescarbeau, R., Ziem, A.,Scrucca, L., Tang, Y., Candan, C., and Hunt., T. (2018).caret: Classification andRegression Training. R package version 6.0-80.
8	Mukherjee, M. and Khushi, M. (2021). Smote-enc: A novel smote-based method to gen-erate synthetic data for nominal and continuous features.Applied System Innovation,4(1):18.
9	Porto, F., de Carvalho Moura, A. M., da Silva, F. C., Bassini, A., Palazzi, D. C., Poltosi,M., de Castro, L. E. V., and Cameron, L. C. (2012). A metaphoric trajectory datawarehouse for olympic athlete follow-up.Concurr. Comput. Pract. Exp., 24(13):1497–1512.
10	Prince, J. and De Vos, M. (2018). A deep learning framework for the remote detection ofparkinson’s disease using smart-phone sensor data. In2018 40th Annual InternationalConference of the IEEE Engineering in Medicine and Biology Society (EMBC), pages3144–3147. IEEE.
11	S. Pereira, R., ferreira da silva, H. M., and A.M Porto, F. (2021).AugmenterR: DataAugmentation for Machine Learning on Tabular Data. R package version 0.1.0.
12	Shorten, C. and Khoshgoftaar, T. M. (2019). A survey on image data augmentation fordeep learning.Journal of Big Data, 6(1):1–48.
13	Sturges, H. A. (1926). The choice of a class interval.Journal of the American StatisticalAssociation, 21(153):65–66.
14	Van Dyk, D. A. and Meng, X.-L. (2001). The art of data augmentation.Journal ofComputational and Graphical Statistics, 10(1):1–50.
15	Vanegas, M. I., Ghilardi, M. F., Kelly, S. P., and Blangero, A. (2018). Machine learningfor eeg-based biomarkers in parkinson’s disease. In2018 IEEE International Confer-ence on Bioinformatics and Biomedicine (BIBM), pages 2661–2665.
16	Zhang, S., Bamakan, S. M. H., Qu, Q., and Li, S. (2019). Learning for personalizedmedicine: A comprehensive review from a deep learning perspective.IEEE Reviewsin Biomedical Engineering, 12:194–208.

Paper Registration

Fill in your paper information

English Information

Authors

Reference