Massively Multilingual Corpus of Sentiment Datasets and Multi-faceted Sentiment Classification Benchmark
This work addresses the problem of culture-dependent sentiment analysis for researchers and practitioners by providing a comprehensive resource and benchmark, though it is incremental in building upon existing datasets.
The authors tackled the challenge of multilingual sentiment analysis by creating the largest open multilingual corpus of sentiment datasets, covering 27 languages from 6 families, and established a benchmark with hundreds of experiments to evaluate models.
Despite impressive advancements in multilingual corpora collection and model training, developing large-scale deployments of multilingual models still presents a significant challenge. This is particularly true for language tasks that are culture-dependent. One such example is the area of multilingual sentiment analysis, where affective markers can be subtle and deeply ensconced in culture. This work presents the most extensive open massively multilingual corpus of datasets for training sentiment models. The corpus consists of 79 manually selected datasets from over 350 datasets reported in the scientific literature based on strict quality criteria. The corpus covers 27 languages representing 6 language families. Datasets can be queried using several linguistic and functional features. In addition, we present a multi-faceted sentiment classification benchmark summarizing hundreds of experiments conducted on different base models, training objectives, dataset collections, and fine-tuning strategies.