Transformer transfer learning emotion detection model: synchronizing socially agreed and self-reported emotions in big data

Abstract

Tactics to determine the emotions of authors of texts such as Twitter messages often rely on multiple annotators who label relatively small data sets of text passages. An alternative method gathers large text databases that contain the authors’ self-reported emotions, to which artificial intelligence, machine learning, and natural language processing tools can be applied. Both approaches have strength and weaknesses. Emotions evaluated by a few human annotators are susceptible to idiosyncratic biases that reflect the characteristics of the annotators. But models based on large, self-reported emotion data sets may overlook subtle, social emotions that human annotators can recognize. In seeking to establish a means to train emotion detection models so that they can achieve good performance in different contexts, the current study proposes a novel transformer transfer learning approach that parallels human development stages: (1) detect emotions reported by the texts’ authors and (2) synchronize the model with social emotions identified in annotator-rated emotion data sets. The analysis, based on a large, novel, self-reported emotion data set (n = 3,654,544) and applied to 10 previously published data sets, shows that the transfer learning emotion model achieves relatively strong performance.

Marketing analytics stages: Demystifying and deploying machine learning

“…Their analysts work with structured and unstructured data by applying advanced techniques, such as Natural Language Processing (NLP) and deep learning. The n-grams approach is a valuable pre-processing solution for NLP (Hartmann et al, 2019; Lee et al, 2023), wherein frequently occurring word combinations are coded quantitatively and non-informative features in natural language data are reduced. This results in coded data that can be analysed; Web Appendix E of Hartmann et al (2019) provides relevant guidelines.…”

“…Recently, transformer models such as BERT and RoBERTa, which are based on deep learning architectures, have been found to outperform other ML techniques in analyzing natural language (Ko et al, 2023; Lee et al, 2023). These models are pre-trained language representation models, as also applied in ChatGPT.…”

Impact of tweets on Indian movies performance: An integrative approach using sentiment analysis and emotion detection

“…Such ML-based systems have their own limitations. In addition, it is currently uncertain how much text and label data is necessary for training models based on self-reported mood data sets.. (Lee 2023) . Social networking sites and microblogging services, such as Facebook and Twitter, provide an unprecedented quantity of (politically pertinent) user-generated content.…”

ML-NLPEmot: Machine Learning-Natural Language Processing Event-Based Emotion Detection Proactive Framework Addressing Mental Health

“…Table I presents the works in the literature which focused on predicting emotions and/or sentiments on social media using machine learning [25], [30]- [37], [40], [41]. This is based on the properties of the considered datasets, the emotion detection algorithms used, and the annotation strategy.…”

“…Most of these works [31]- [37] focused on detecting the sentiments (negative, neutral, and/or positive). The rest of the works focused on a few emotions, mainly Fear, Sadness and Anger [25], [30], [31], [40], [41], [44], [45].…”