The Unreasonable Effectiveness of Machine Learning in Moldavian versus Romanian Dialect Identification
This work addresses the problem of understanding and improving dialect identification for NLP researchers, but it is incremental as it builds on a prior shared task.
The paper investigates why machine learning models achieve high accuracy in distinguishing Moldavian and Romanian dialects, with top models scoring 0.895 macro F1, and finds that models maintain performance even at sentence level and across domains like tweets, while humans perform worse.
Motivated by the seemingly high accuracy levels of machine learning models in Moldavian versus Romanian dialect identification and the increasing research interest on this topic, we provide a follow-up on the Moldavian versus Romanian Cross-Dialect Topic Identification (MRC) shared task of the VarDial 2019 Evaluation Campaign. The shared task included two sub-task types: one that consisted in discriminating between the Moldavian and Romanian dialects and one that consisted in classifying documents by topic across the two dialects of Romanian. Participants achieved impressive scores, e.g. the top model for Moldavian versus Romanian dialect identification obtained a macro F1 score of 0.895. We conduct a subjective evaluation by human annotators, showing that humans attain much lower accuracy rates compared to machine learning (ML) models. Hence, it remains unclear why the methods proposed by participants attain such high accuracy rates. Our goal is to understand (i) why the proposed methods work so well (by visualizing the discriminative features) and (ii) to what extent these methods can keep their high accuracy levels, e.g. when we shorten the text samples to single sentences or when we use tweets at inference time. A secondary goal of our work is to propose an improved ML model using ensemble learning. Our experiments show that ML models can accurately identify the dialects, even at the sentence level and across different domains (news articles versus tweets). We also analyze the most discriminative features of the best performing models, providing some explanations behind the decisions taken by these models. Interestingly, we learn new dialectal patterns previously unknown to us or to our human annotators. Furthermore, we conduct experiments showing that the machine learning performance on the MRC shared task can be improved through an ensemble based on stacking.