Ensemble and Incremental Learning for Norm Violation Detection

The use of norms to guide and coordinate interactions has gained tremendous attention in the multiagent community. However, as the interest moves towards dynamic socio-technical systems, where human and software agents interact and interactions are required to adapt to the human's changing needs, new challenges arise. For instance, different agents (human or software) might not have the same understanding of what it means to violate a norm (e.g., what characterizes hate speech), or that their understanding of a norm might change over time (e.g., what constitutes an acceptable response time). The challenge is to address these issues by learning the meaning of a norm violation from the limited interaction data. For this, we propose a framework that learns and updates the meaning of a norm violation from interaction data through a combination of ensemble and incremental learning techniques. Ensemble learning handles the imbalance class distribution of the interaction stream, and incremental learning is used to continuously update the ensemble models as community members interact, which is an essential feature to keep the ensemble models in accordance with the latest community view on the meaning of norm violation. We evaluate the proposed approach in the use case of Wikipedia article edits, where interactions revolve around editing articles and the norm in question is prohibiting vandalism. Results show that the proposed framework can learn the meaning of a norm violation in a setting with data imbalance and concept drift.