Mondrian Forest for Data Stream Classification Under Memory Constraints

TL;DR

This paper adapts the online Mondrian forest algorithm for data stream classification to operate under strict memory constraints, introducing strategies for updating and trimming trees to handle concept drift and limited memory.

Contribution

It proposes five out-of-memory strategies and trimming mechanisms for Mondrian trees, enabling effective data stream classification on memory-limited devices.

Findings

Extend Node strategy is the best out-of-memory approach across scenarios.

Trimming mechanisms improve robustness to concept drift.

Methods are implemented in the open-source OrpailleCC library.

Abstract

Supervised learning algorithms generally assume the availability of enough memory to store their data model during the training and test phases. However, in the Internet of Things, this assumption is unrealistic when data comes in the form of infinite data streams, or when learning algorithms are deployed on devices with reduced amounts of memory. In this paper, we adapt the online Mondrian forest classification algorithm to work with memory constraints on data streams. In particular, we design five out-of-memory strategies to update Mondrian trees with new data points when the memory limit is reached. Moreover, we design trimming mechanisms to make Mondrian trees more robust to concept drifts under memory constraints. We evaluate our algorithms on a variety of real and simulated datasets, and we conclude with recommendations on their use in different situations: the Extend Node strategy appears as the best out-of-memory strategy in all configurations, whereas different trimming mechanisms should be adopted depending on whether a concept drift is expected. All our methods are implemented in the OrpailleCC open-source library and are ready to be used on embedded systems and connected objects.