Abstract
In this paper, we address the task of zero-shot cross-lingual news sentiment classification. Given the annotated dataset of positive, neutral, and negative news in Slovene, the aim is to develop a news classification system that assigns the sentiment category not only to Slovene news, but to news in another language without any training data required. Our system is based on the multilingual BERTmodel, while we test different approaches for handling long documents and propose a novel technique for sentiment enrichment of the BERT model as an intermediate training step. With the proposed approach, we achieve state-of-the-art performance on the sentiment analysis task on Slovenian news. We evaluate the zero-shot cross-lingual capabilities of our system on a novel news sentiment test set in Croatian. The results show that the cross-lingual approach also largely outperforms the majority classifier, as well as all settings without sentiment enrichment in pre-training.
Highlights
Sentiment analysis is one of the most popular applications of natural language processing (NLP) and has found many areas of applications in customers’ product reviews, survey textual responses, social media, etc
We present our approach to cross-lingual news sentiment analysis, where given an available sentiment-annotated dataset of news in Slovene [3], we propose a news sentiment classification model for other languages
We focus on Croatian, where the news dataset is provided by 24sata, one of the leading portals in Croatia, and was labeled with the same sentiment annotation scheme as the Slovenian dataset in order to allow comparison in a zero-shot learning setting where no annotations in the target language are expected
Summary
Sentiment analysis is one of the most popular applications of natural language processing (NLP) and has found many areas of applications in customers’ product reviews, survey textual responses, social media, etc. It analyzes users’ opinions on various topics, such as politics, health, education, etc. Bowden et al [9] took a step further and tried to improve the forecasting of three economic variables, inflation, output growth, and unemployment, via sentiment modeling They concluded that, using sentiment analysis, out of the three variables observed, the forecasting can be effectively improved for unemployment
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
