An Automatic Unlabeled Selection for CO-training REGressors (AU-COREG)

Main Article Content

Sirikwan Kheereesuwannakul
Eakasit Pacharawongsakda

Abstract

This research aims to improve the performance of semi-supervised learning by automatically select unlabeled data. The proposed method uses two regression models to estimate values for unlabeled data, then cluster the data into groups. Therefore, similar data are assigned in the same group and the different data are assigned into the different groups. After that, the method selects each group representative that have least error and append into training data. Then, we repeat until we have enough training data. From experimental results with three datasets, we found that the proposed method can improve performance and reduce computation time by 84%, comparing to previous work.

Downloads

Download data is not yet available.

Article Details

How to Cite
[1]
S. Kheereesuwannakul and E. Pacharawongsakda, “An Automatic Unlabeled Selection for CO-training REGressors (AU-COREG)”, JIST, vol. 10, no. 1, pp. 10-23, Jun. 2020.
Section
Research Article: Soft Computing (Detail in Scope of Journal)

References

[1] Bizibl Marketing, “10 Key Marketing Trends for 2017 and Ideas for Exceeding Customer Expectations,” Bizibl Marketing, June 16, 2019. [Online]. Available: https://bizibl.com/marketing/download/10-key-marketing-trends-2017-and-ideas-exceeding-customer-expectations. [Accessed: June 16, 2020].

[2] Blum A., Mitchell T., “Combining labeled and unlabeled data with co-training,” COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory, July, 1998, pp. 92-100.

[3] Didaci L., Fumera, G., Roli, F., “Analysis of co-training algorithm with very small training sets,” Gimel’farb, G., et al. (eds.) SSPR/SPR 2012. LNCS., Springer, Heidelberg., vol. 7626, 2012.

[4] R. Wang and L. Li, "The performance improvement algorithm of co-training by committee," 2016 5th International Conference on Computer Science and Network Technology (ICCSNT), Changchun, 2016, pp. 407-412, doi: 10.1109/ICCSNT.2016.8070190.

[5] Sousa R., Gama J., “Co-training Semi-Supervised Learning for Single-Target Regression in Data Streams Using AMRules,” In: Kryszkiewicz M., Appice A., Ślęzak D., Rybinski H., Skowron A., Raś Z. (eds) Foundations of Intelligent Systems, ISMIS 2017, Lecture Notes in Computer Science, Vol. 10352, 2017.

[6] F Ma, D Meng, Q Xie, Z Li, X Don, “Self-paced co-training,” Proceedings of the 34th International Conference on Machine Learning, Vol. 70, pp. 2275-2284, 2017.

[7] Zhi Hua., Ming Li., “Semi-supervised regression with co-training,” IJCAI’05 proceeding of the 19th international joint conference on artificial intelligence, July, 2005, pp. 908–913.