Emotion Analysis on Twitter: The Hidden Challenge

Emotion Analysis on Twitter: the Hidden Challenge Luca Dini, André Bittar Holmes Semantic Solutions Grenoble, France E-mail: dini@ho2s.com, bittar@ho2s.com In this paper, we present an experiment to detect emotions in tweets. Unlike much previous research, we draw the important distinction between the tasks of emotion detection in a closed world assumption (i.e. every tweet is emotional) and the complicated task of identifying emotional versus non-emotional tweets. Given an apparent lack of appropriately annotated data, we created two corpora for these tasks. We describe two systems, one symbolic and one based on machine learning, which we evaluated on our datasets. Our evaluation shows that a machine learning classifier performs best on emotion detection, while a symbolic approach is better for identifying relevant (i.e. emotional) tweets. Keywords: emotion analysis, Twitter, corpora 1. Introduction In recent years so-called “emotional marketing” has become a key factor of success for many B2C companies, especially global ones. Put simply, emotional marketing has the basic goal of convincing customers that a brand or a product are not just brands or products, but that they are imbued with qualities that favor an emotional response or attachment in consumers. Any emotional marketing strategy starts with an in-depth understanding of customers’ emotional or motivational drivers, and this is achieved either via open question surveys or by analyzing emotional reactions on social media. This paper has the goal of validating the feasibility of automatic emotional analysis on social media. It should be noticed that standard sentiment analysis, by which we mean the detection of opinions as positive versus negative, or possibly ranked on a scale of intensity, is of little use for our purposes, both practically and conceptually. Practically, the fact that a customer expresses, for instance, a positive opinion has no impact on an emotional evaluation. For example, the sentences I love this cup and This cup is very good have an identical opinion value, but definitely a different emotional tonality. Conceptually, whereas opinions are usually expressed in a quite explicit way, the language of emotions is much more difficult to interpret as it relies on indirect signals. We emphasize the difficulty of the task at hand by comparing a symbolic approach to emotion analysis and one based on machine learning. We show that while the task of classifying emotions in a closed world assumption (i.e. each text has an emotional value) is a relatively easy task, the real challenge lies in distinguishing emotional texts from non-emotional ones (information filtering). We also show that a statistical approach outperforms a symbolic one in emotion classification, while the contrary is true when considering information filtering. Concretely, we focused on two different tasks. The first is the categorization of tweets according to the emotion they express – a single emotion per tweet, from a set of 6 basic emotions. As we explain in detail in Erreur ! Source du renvoi introuvable., currently available corpora do not, to our knowledge, provide appropriate data for evaluating the automation of this task and a major objective …

Content truncated to 500 words