This paper presents a comprehensive benchmark and survey of fully unsupervised concept drift detectors (UCDD) designed to identify and adapt to concept drift in real-world data streams. Concept drift refers to the phenomenon where the statistical properties of a data stream change over time, leading to the deterioration of model accuracy if not detected and adjusted. The study reviews the state of the art in UCDDs, evaluates their performance on various real-world datasets, and identifies challenges and open research areas in the field. Through empirical experiments and a systematic review of existing methods, we highlight key factors influencing the performance of these detectors in unsupervised environments.

Unsupervised Drift Detection, Data Streams, Concept Drift, Machine Learning, Adaptive Learning, Streaming Data Analysis

Gama, J., et al. (2014). A survey on concept drift adaptation. ACM Computing Surveys (CSUR), 46(4), 1-36.

Bifet, A., & Gama, J. (2010). Learning from time-changing data with adaptive windowing. In Proceedings of the 2007 SIAM International Conference on Data Mining (pp. 443-448).

Hoeffding, W. (1963). Probability inequalities for sums of bounded random variables. Journal of the American Statistical Association, 58(301), 13-30.

Bifet, A., Holmes, G., & Gama, J. (2010). Data stream mining: A practical approach. Springer.

Hinkley, D. V. (1970). Inferences about the change-point in a sequence of random variables. Biometrika, 57(1), 1-17.

Kifer, D., Gehrke, J., & Ramakrishnan, R. (2004). Detecting change in data streams. In Proceedings of the 30th International Conference on Very Large Data Bases (pp. 180-191).

López, E. V., & Bifet, A. (2017). An overview of concept drift detection and learning in data streams. In Proceedings of the International Workshop on Knowledge Discovery in Data Streams (pp. 39-54). Springer.

Gama, J., Zighed, D. A., & Žliobaite, I. (2014). Concept drift: A review and research directions. Data Mining and Knowledge Discovery, 28(1), 77-95.

Zhang, M., & Wang, X. (2015). A density-based approach for concept drift detection in data streams. Expert Systems with Applications, 42(3), 1047-1061.

Hodge, V. J., & Austin, J. (2004). A survey of outlier detection methodologies. Artificial Intelligence Review, 22(2), 85-126.

Gama, J., et al. (2015). Framework for evaluating concept drift detection algorithms. Proceedings of the 18th European Conference on Artificial Intelligence (ECAI 2015).

Schlimmer, J. C., & Granger, R. H. (1986). Incremental learning from noisy data. Machine Learning, 1(1), 15-34.

Zliobaite, I., & Gama, J. (2016). A survey on concept drift adaptation. ACM Computing Surveys (CSUR), 48(2), 1-39.

Bifet, A., & Gavaldà, R. (2007). Learning from time-changing data with adaptive windowing. Proceedings of the 2007 SIAM International Conference on Data Mining (pp. 443-448).

Friedman, J., Hastie, T., & Tibshirani, R. (2001). The elements of statistical learning: Data mining, inference, and prediction. Springer Science & Business Media.

Zhou, Z. H., & Li, M. (2010). Semi-supervised learning by embedding model selection. IEEE Transactions on Knowledge and Data Engineering, 22(3), 410-421.

Nguyen, L., & Widmer, G. (2018). Unsupervised change detection in data streams. Data Mining and Knowledge Discovery, 32(6), 1461-1483.

Zhang, M., & Wu, L. (2013). Concept drift detection in data streams using a density-based approach. Journal of Machine Learning Research, 14(1), 2543-2574.

Gama, J., & Rocha, R. (2013). Drift detection in data streams with sudden and incremental changes. Proceedings of the 16th International Conference on Artificial Intelligence and Statistics (AISTATS 2013).

Li, B., & Li, J. (2019). Unsupervised concept drift detection in real-world streams. IEEE Transactions on Neural Networks and Learning Systems, 30(4), 1164-1175.