Concept Drift - Possible Remedies

Possible Remedies

To prevent deterioration in prediction accuracy because of concept drift, both active and passive solutions can be adopted. Active solutions rely on triggering mechanisms, e.g., change-detection tests (Basseville and Nikiforov 1993; Alippi and Roveri, 2007) to explicitly detect concept drift as a change in the statistics of the data-generating process. In stationary conditions, any fresh information made available can be integrated to improve the model. Differently, when concept drift is detected, the current model is no more up-to-date and must be substituted with a new one to maintain the prediction accuracy (Gama et al., 2004; Alippi et al., 2011). On the contrary, in passive solutions the model is continuously updated, e.g., by retraining the model on the most recently observed samples (Widmer and Kubat, 1996), or enforcing an ensemble of classifiers (Elwell and Polikar 2011).

Contextual information, when available, can be used to better explain the causes of the concept drift: for instance, in the sales prediction application, concept drift might be compensated by adding information about the season to the model. By providing information about the time of the year, the rate of deterioration of your model is likely to decrease, concept drift is unlikely to be eliminated altogether. This is because actual shopping behavior does not follow any static, finite model. New factors may arise at any time that influence shopping behavior, the influence of the known factors or their interactions may change.

Concept drift cannot be avoided for complex phenomenon that are not governed by fixed laws of nature. All processes that arise from human activity, such as socioeconomic processes, and biological processes are likely to experience concept drift.Therefore periodic retraining, also known as refreshing, of any model is necessary.