Abstract
Dynamics are fundamental properties of batch learning processes. Recently, monitoring dynamic processes has interested many researchers due to the importance of dealing with time-changing data stream processes in real-world applications. In this article, a dynamic weighted majority (DWM)-based identification model is proposed for monitoring small, large as well as covariate shifts in nonstationary processes. The proposed method applies DWM ensemble method to aggregate decisions of different control charts to improve single charts’ performances and to reduce the risk of choosing a nonadequate chart. Also in order to improve the shift adaptation mode, a prediction of class label is used to help in classifying the shift during the changing of the process toward the approximated right direction. The new proposed ensemble chart has the ability to deal with complex datasets and presents a concrete shift identification method based on a classification learning technique of changes in nonstationary processes.
Similar content being viewed by others
References
Asensio AS, Puig AO, Golobardes E (2014) Robust on line neural learning classifier system for data stream classification tasks. J Soft Comput 18(8):1441–1461
Butz MV (1995) Rule-based evolutionary online learning systems a principal approach to LCS analysis and design, studies in fuzziness and soft computing. Springer, Berlin
Deckert M (2011) Batch weighted ensemble for mining data streams with concept drift. In: Proceedings of the 19th international symposium, foundations of intelligent systems, Poland, pp 290–299
Díaz AO, del Campo-Ávila J, Ramos-Jiménez G, Blanco IF, Mota YC, Hechavarría AM, Morales-Bueno R (2014) Fast adapting ensemble: a new algorithm for mining data streams with concept drift. The Scientific World Journal, Article ID 235810 (in press)
Flaig JJ (2014) The Shewhart EWMA automatic control chart. Glob J Res Eng 14(1):21–26
Flaig JJ (2014) Construction of combined charts based on combining functions. Appl Math Sci 8(84):4187–4200
Gama J, Medas P, Castillo G, Rodrigues P (2004) Learning with drift detection. In: SBIA Brazilian symposium on artificial intelligence, Springer Verlag, pp 286–295
Gibbons RD (1999) Use of combined Shewhart CUSUM control charts for ground water monitoring applications. Gr Water 37(5):682–691
Kolter JZ, Maloof MA (2005) Using additive expert ensembles to cope with concept drift. In: Proceedings of the 22nd international conference on machine learning (ICML 2005), Bonn, Germany, ACM Press, pp 449–456
Kolter JZ, Maloof MA (2007) Dynamic weighted majority: an ensemble method for drifting concepts. J Mach Learn Res 8:2755–2790
Liu S, Yamada M, Collier N, Sugiyama M (2013) Change point detection in time series data by relative density ratio estimation. J Neural Netw 43:72–83
Li PP, Wu X, Hu X (2010) Mining recurring concept drifts with limited labeled streaming data. In: Proceedings of the 2nd Asian conference on machine learning, ACML, Tokyo, Japan, November 8–10, pp 241–252
Lucas James M (1982) Combined Shewhart CUSUM quality control schemes. J Qual Technol 14(2):51–59
Maloof MA, Michalski RS (2004) Incremental learning with partial instance memory. Artif Intell 154:95–126
Mejri D, Limam M, Weihs C (2015) Monitoring a dynamic weighted majority method based on real datasets with concept drift, Springer International Publishing Switzerland, Wilhelm A (ed.), Analysis of large and complex data, studies in classification, data analysis, and knowledge organization, doi:10.1007/978-3-319-25226-1-21
Mejri D, Khanchel R, Limam M (2013) An ensemble method for concept drift in nonstationary environment. J Stat Comput Simul 83:1115–1128
Michalski R (1969) On the quasi minimal solution of the general covering problem. In: Proceedings of the fifth international symposium on information processing, pp 125–128
Nishida K, Yamauchi K, Omori T (2005) ACE: adaptive classifiers ensemble system for concept drifting environments. In: Multiple classifier systems, \(6{th}\) international workshop, MCS 2005, Seaside, CA, USA, June 13–15, Proceedings, pp 176–185
Raza H, Prasad G, Li Y (2013) Dataset shift detection in nonstationary environments using EWMA charts. In: Proceedings of the IEEE international conference on systems, man and cybernetics, pp 3151–3156
Raza H, Prasad G, Li Y (2015) EWMA model based shift detection methods for detecting covariate shifts in non stationary environments. Pattern Recognit 48(3):659–669
Sampaio ES, Ho LL, de Medeiros PG (2014) A combined \(np_{x}\) \(\overline{X}\) Control Chart to monitor the process mean in Two Stage Sampling. Qual Reliab Eng Int 30(7):1003–1013
Street WN, Kim Y (2001) A streaming ensemble algorithm (sea) for large-scale classification. In: Proceedings of the seventh ACM SIGKDD international conference on knowledge discovery and data mining, KDD 01, New York, NY, USA, ACM, pp 377–382
Wang H, Fan W, Yu PS, Han J (2003) Mining concept-drifting data streams using ensemble classifiers, In Proceedings of the ninth ACM SIGKDD international conference on knowledge discovery and data mining, KDD 03, New York, NY, USA. ACM, pp 226–235
Wilson SW (1995) Classifier fitness based on accuracy. Evolut Comput 3(2):149–175
Acknowledgments
This study was funded by the Deutsher Akademischer Austauschdienst (DAAD) with the Grant No. 91526665 and by a research employment at the Technische Universität Dortmund.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Authors declare that they have no conflict of interest.
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Additional information
Communicated by V. Loia.
Rights and permissions
About this article
Cite this article
Mejri, D., Limam, M. & Weihs, C. A new dynamic weighted majority control chart for data streams. Soft Comput 22, 511–522 (2018). https://doi.org/10.1007/s00500-016-2351-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-016-2351-3