Candidate: Liuyan Chen
Title: Text Mining to Understand Emotion Triggers
Date: April 5, 2019
Time: 11:30 am
Place: CPH 2371
Supervisor(s): Golab, Lukasz
Abstract:
In computational linguistics, most sentiment analysis builds binary classification models on customer reviews data to predict whether a review is positive or negative. In this thesis, we go a step further and build interpretable classification models to predict the emotion associated with the text (such as happy, sad, productive and tired). This analysis is enabled by a unique journaling dataset containing short pieces of text and the associated emotional status self-reported by the writer. To further study what people feel emotional about (emotion triggers), we perform model interpretation.
We make two main contributions. First, we apply state-of-the-art text mining methodologies to extract emotion triggers from text, during which we discover and solve an issue of the attention mechanism in a popular deep learning model (Dynamic Memory Network (DMN)). Second, we obtain data-driven evidence of emotion triggers, which can help the emotion trigger identification process in emotion regulation therapy.