Empathy detection from textual data is a complex task that requires an understanding of both the content and context of the text. This study presents a BERT-based context-aware approach to enhance empathy detection in conversations and essays. We participated in the WASSA 2024 Shared Task, focusing on two tracks: empathy and emotion prediction in conversations (CONV-turn) and empathy and distress prediction in essays (EMP). Our approach leverages contextual information by incorporating related articles and emotional characteristics as additional inputs, using BERT-based Siamese (parallel) architecture. Our experiments demonstrated that using article summaries as context significantly improves performance, with the parallel BERT approach outperforming the traditional method of concatenating inputs with the ‘[SEP]‘ token. These findings highlight the importance of context-awareness in empathy detection and pave the way for future improvements in the sensitivity and accuracy of such systems.

Various recent studies show that the performance of named entity recognition (NER) systems developed for well-formed text types drops significantly when applied to tweets. The only existing study for the highly inflected agglutinative language Turkish reports a drop in F-Measure from 91% to 19% when ported from news articles to tweets. In this study, we present a new named entity-annotated tweet corpus and a detailed analysis of the various tweet-specific linguistic phenomena. We perform comparative NER experiments with a rule-based multilingual NER system adapted to Turkish on three corpora: a news corpus, our new tweet corpus, and another tweet corpus. Based on the analysis and the experimentation results, we suggest system features required to improve NER results for social media like Twitter.

This paper presents an evaluation of the use of machine translation to obtain and employ data for training multilingual sentiment classifiers. We show that the use of machine translated data obtained similar results as the use of native-speaker translations of the same data. Additionally, our evaluations pinpoint to the fact that the use of multilingual data, including that obtained through machine translation, leads to improved results in sentiment classification. Finally, we show that the performance of the sentiment classifiers built on machine translated data can be improved using original data from the target language and that even a small amount of such texts can lead to significant growth in the classification performance.