In dynamic environments, predictive models face the challenge of concept drift - changes in data distributions that degrade model performance over time. Concept drift detection and visualization can bring valuable insight into the data dynamics, especially for multidimensional data, and are related to visual knowledge discovery. To address this, we introduce a novel visualization model based on parallel coordinates, Parallel Histograms through Time (PHT). It allows visualization and identification of concept drift in the input data, enabling the simultaneous observation of the evolution of several features. When paired with a classifier, the PHT model provides a simultaneous visualization of the data changes in terms of (i) histograms over time and (ii) variation of the characteristics of the features. Although a classifier can identify changes, the PHT model enhances this capability by visually representing the detected drift, offering good premises for a comprehensive justification of the observed changes. Experimental evaluations on synthetic (CIRCLES, SINE1) and real-world (WEATHER, ELECTRICITY) datasets demonstrate that our method detects drift with high accuracy. In conjunction with domain-specific causality analysis, our model can be used to justify the identified and visualized concept drifts.
GamaJ.Knowledge discovery from data streams. 1st ed.Chapman Hall/CRC, 2010.
2.
HinderFVaquetVHammerB. One or two things we know about concept drift–a survey on monitoring evolving environments. Front Artif Intell 2024. DOI: 10.3389/frai.2024.1330257.
3.
Suárez-CetruloALQuintanaDCervantesA.A survey on machine learning for recurring concept drifting data streams. Expert Syst Appl2023; 213: 118934.
4.
Giraud-CarrierC.A note on the utility of incremental learning. AI Commun2000; 13(4): 215–223.
5.
GâlmeanuHAndonieR. Incremental/decremental SVM for function approximation. In: 2008 11th international conference on optimization of electrical and electronic equipment, 2008, pp.155–160. IEEE.
6.
GâlmeanuHAndonieR.Concept drift adaptation with incremental–decremental SVM. Appl Sci2021; 11(20): 97–106.
7.
LuJLiuADongF, et al. Learning under concept drift: a review. IEEE Trans Knowl Data Eng2018; 31(12): 2346–2363.
8.
KovalerchukBAndonieRDatiaN, et al. Visual knowledge discovery with artificial intelligence: challenges and future directions. Chapter 1. Springer, 2022, pp.1–27.
9.
KovalerchukBNazemiKAndonieR, et al. Integrating artificial intelligence and visualization for visual knowledge discovery. vol. 1014. Springer, 2022.
10.
KovalerchukBNazemiKAndonieR, et al. Artificial intelligence and visualization: advancing visual knowledge discovery. Springer, 2024.
11.
AgrahariSSinghAK.Comparison based analysis of window approach for concept drift detection and adaptation. Appl Intell2025; 55(1): 1–17.
12.
FewS. Now you see it: simple visualization techniques for quantitative analysis. 1st ed. Analytics Press, 2009.
13.
HinderFVaquetVHammerB.Model-based explanations of concept drift. Neurocomputing2023; 555: 126640.
14.
PrattKBTschapekG. Visualizing concept drift. In: Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD’03, 2003, pp.735–740. New York, NY: Association for Computing Machinery. DOI:10.1145/956750.956849.
15.
SchneiderBKeimDAEl-AssadyM. Datashiftexplorer: visualizing and comparing change in multidimensional data for supervised learning. In: Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - IVAPP, 2020, pp. 141–148. INSTICC, SciTePress. DOI: 10.5220/0008940801410148.
16.
WangXChenWXiaJ, et al. Conceptexplorer: visual analysis of concept drifts in multi-source time-series data. In: 2020 IEEE Conference on Visual Analytics Science and Technology (VAST), 2020, pp.1–11. Los Alamitos, CA: IEEE Computer Society. DOI: 10.1109/VAST50239.2020.00006.
17.
WebbGILeeLKGoethalsB, et al. Analyzing concept drift and shift from sample data. Data Min Knowl Discov2018; 32: 1179–1199.
18.
GâlmeanuHAndonieR.Weighted incremental-decremental support vector machines for concept drift with shifting window. Neural Networks2022; 152: 528–541.
19.
GâlmeanuHAndonieR. Concept drift visualization of SVM with shifting window. In: 2024 28th International Conference Information Visualisation (IV), 2024, pp.1–7. Los Alamitos, CA: IEEE Computer Society. DOI: 10.1109/IV64223.2024.00046.
20.
BifetAGavaldàR. Learning from time-changing data with adaptive windowing. In: Proceedings of the Seventh SIAM International Conference on Data Mining, Minneapolis, USA, SDM’07, vol. 7, 2007, p.6. DOI: 10.1137/1.9781611972771.42.
21.
TsymbalA.The problem of concept drift: definitions and related work. Technical report, Department of Computer Science, Trinity College, Dublin, 2004.
22.
IwashitaASPapaJP.An overview on concept drift learning. IEEE Access2019; 7: 1532–1547.
23.
PesaranghaderA. The tornado framework for data stream mining (Python implementation). Technical report, 2020. https://github.com/alipsgh/tornado (accessed July 1, 2025).
24.
CarpenterGAGrossbergS.The art of adaptive pattern recognition by a self-organizing neural network. Computer1988; 21(3): 77–88.
25.
GuoHLiHRenQ, et al. Concept drift type identification based on multi-sliding windows. Inf Sci2022; 585: 1–23.
26.
KabirMABegumSAhmedMU, et al. Code: a moving-window-based framework for detecting concept drift in software defect prediction. Symmetry2022; 14(12): 2508.
27.
YinYLiPChenJ. A variable sliding window algorithm based on concept drift for frequent pattern mining over data streams. In: 2022 IEEE 28th international conference on parallel and distributed systems (ICPADS), 2023, pp.818–825. DOI: 10.1109/ICPADS56603.2022.00111.
28.
CauwenberghsGPoggioT.Incremental and decremental support vector machine learning. In: Proceedings of the 13th international conference on neural information processing systems, NIPS’00, 2000, pp.388–394. Cambridge, MA: MIT Press.
29.
GâlmeanuHAndonieR. Implementation issues of an incremental and decremental SVM. In: Proceedings of the 18th international conference on artificial neural networks, Part I, ICANN ’08, 2008, pp.325–335. Berlin, Heidelberg: Springer-Verlag. DOI: 10.1007/978-3-540-87536-9.34.
30.
SyedNALiuHSungKK.Handling concept drifts in incremental learning with support vector machines. In: Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD’99, 1999, pp.317–321. New York, NY: Association for Computing Machinery. DOI: 10.1145/312129.312267.
31.
RüpingS. Incremental learning with support vector machines. In: Proceedings 2001 IEEE international conference on data mining, 2001, pp.641–642.
32.
ZareMoodiPSiahroudiSKBeigyH. A support vector based approach for classification beyond the learned label space in data streams. In: Proceedings of the 31st annual ACM symposium on applied computing, SAC’16, 2016, pp.910–915. New York, NY: Association for Computing Machinery. DOI: 10.1145/2851613.2851652.
33.
JainMKaurGSaxenaV.A k-means clustering and SVM based hybrid concept drift detection technique for network anomaly detection. Expert Syst Appl2022; 193: 116510.
34.
ChitrakarRHuangC.Selection of candidate support vectors in incremental SVM for network intrusion detection. Comput Secur2014; 45: 231–241.
35.
ChenYXiongJXuW, et al. A novel online incremental and decremental learning algorithm based on variable support vector machine. Cluster Comput2019; 22: 7435–7445.
36.
KlinkenbergRJoachimsT.Detecting concept drift with support vector machines. In: ICML’00: proceedings of the Seventeenth International Conference on Machine Learning, ICML, 2000, pp.487–494.
37.
KlinkenbergR.Learning drifting concepts: example selection vs. example weighting. Intell Data Anal2004; 8(3): 281–300.
38.
BlackardJADeanDJ.Comparative accuracies of artificial neural networks and discriminant analysis in predicting forest cover types from cartographic variables. Comput Electron Agric1999; 24(3): 131–151.
39.
GamaJŽliobaitėIBifetA, et al. A survey on concept drift adaptation. ACM Comput Surv2014; 46(4): 1–37.
40.
PesaranghaderAViktorHPaquetE.Reservoir of diverse adaptive learners and stacking fast Hoeffding drift detection methods for evolving data streams. Mach Learn2018; 107: 1711–1743.
