Performance Evaluation of Feature Selection Algorithms Applied to Online Learning in Concept Drift Environments

• Matheus B. de Moraes UNICAMP
• André L. S. Gradvohl UNICAMP

DOI: https://doi.org/10.5753/eniac.2018.4438

Resumo

Data streams are transmitted at high speeds with huge volume and may contain critical information need processing in real-time. Hence, to reduce computational cost and time, the system may apply a feature selection algorithm. However, this is not a trivial task due to the concept drift. In this work, we show that two feature selection algorithms, Information Gain and Online Feature Selection, present lower performance when compared to classification tasks without feature selection. Both algorithms presented more relevant results in one distinct scenario each, showing final accuracies up to 14% higher. The experiments using both real and artificial datasets present a potential for using these methods due to their better adaptability in some concept drift situations.

Referências

Andrade, H., Gedik, B., and Turaga, D. (2011). Fundamentals of stream processing: Application design, systems, and analytics. Fundamentals of Stream Processing: Application Design, Systems, and Analytics, pages 1–529.

Bifet, A., Holmes, G., Kirkby, R., and Pfahringer, B. (2010). MOA: massive online analysis. Journal of Machine Learning Research, 11:1601–1604.

Bifet, A. and Kirkby, R. (2011). Data Stream Mining. Methodology, 8(May):127–141.

de Moraes, M. B. (2018). Data StreamsWith Feature Selection. Disponível em: https://github.com/mbdemoraes/data-streams-feature-selection/. Last accessed in May 11th, 2018.

Demsar, J. (2006). Statistical Comparisons of Classifiers over Multiple Data Sets. Journal of Machine Learning Research, 7:1–30.

Gama, J., Zliobait˙e, I., Bifet, A., Pechenizkiy, M., and Bouchachia, A. (2014). A survey on concept drift adaptation. ACM Computing Surveys, 46(4):1–37.

Gradvohl, A. L. S. (2016). Investigating metrics to build a benchmark tool for complex event processing systems. Proceedings - 2016 4th International Conference on Future Internet of Things and Cloud Workshops, W-FiCloud 2016, pages 143–147.

Gradvohl, A. L. S., Senger, H., Arantes, L., and Sens, P. (2014). Comparing distributed online stream processing systems considering fault tolerance issues. Journal of Emerging Technologies in Web Intelligence, 6(2):174–179.

Han, J. and Kamber, M. (2006). Data Mining Concepts and Techniques, volume 2. Elsevier, Burlington.

Jankowski, D., Jackowski, K., and Cyganek, B. (2016). Learning Decision Trees from Data Streams with Concept Drift. Procedia Computer Science, 80:1682–1691.

Katakis, I., Tsoumakas, G., Banos, E., Bassiliades, N., and Vlahavas, I. (2009). An adaptive personalized news dissemination system. Journal of Intelligent Information Systems, 32(2):191–212.

Katakis, I. and Tsoumakas, I. V. (2005). On the utility of incremental feature selection for the classification of textual data streams. In Fagerberg, J., Mowery, D. C., and Nelson,

R. R., editors, Advances in Informatics, pages 338–348.

SpringerLink, Volos,Greece. March, J. G. (1991). Exploracion Y Explotacion En Aprendizaje Organizacional.

Quinlan, J. R. (1986). Induction of Decision Trees. Machine Learning, 1(1):81–106.

Ramírez-Gallego, S., Krawczyk, B., García, S., Wo´zniak, M., and Herrera, F. (2017). A survey on data preprocessing for data stream mining: Current status and future directions. Neurocomputing, 239:39–57.

Turkov, P., Krasotkina, O., Mottl, V., and Sychugov, A. (2016). Feature Selection for Handling Concept Drift in the Data Stream Classification. In 12th International Conference on Machine Learning and Data Mining in Pattern Recognition, pages 614–629. SpringerLink, New York, USA.

Wang, J., Zhao, P., Hoi, S. C., and Jin, R. (2014). Online feature selection and its applications. IEEE Transactions on Knowledge and Data Engineering, 26(3):698–710.