# Leveraging Large Amounts of Weakly Supervised Data for Multi-Language Sentiment Classification

## TL;DR

This paper introduces a method for multi-lingual sentiment classification that leverages large weakly-supervised datasets and pre-training, achieving state-of-the-art results without relying on English-language supervision.

## Contribution

The authors propose a novel multi-lingual sentiment classification approach that does not require English supervision and demonstrates effective use of weakly-supervised data and pre-training.

## Key findings

- Achieved state-of-the-art performance on SemEval-2016 sentiment benchmark.
- Multi-language model generalizes well across languages.
- Pre-training significantly improves classification accuracy.

## Abstract

This paper presents a novel approach for multi-lingual sentiment classification in short texts. This is a challenging task as the amount of training data in languages other than English is very limited. Previously proposed multi-lingual approaches typically require to establish a correspondence to English for which powerful classifiers are already available. In contrast, our method does not require such supervision. We leverage large amounts of weakly-supervised data in various languages to train a multi-layer convolutional network and demonstrate the importance of using pre-training of such networks. We thoroughly evaluate our approach on various multi-lingual datasets, including the recent SemEval-2016 sentiment prediction benchmark (Task 4), where we achieved state-of-the-art performance. We also compare the performance of our model trained individually for each language to a variant trained for all languages at once. We show that the latter model reaches slightly worse - but still acceptable - performance when compared to the single language model, while benefiting from better generalization properties across languages.