GradAscent at EmoInt-2017: Character- and Word-Level Recurrent Neural Network Models for Tweet Emotion Intensity Detection
This work addresses emotion intensity detection in noisy, short tweets, which is important for applications in sentiment analysis and social media monitoring, but it is incremental as it builds on existing neural network and lexicon-based approaches.
The paper tackled the problem of predicting emotion intensity values in tweets for four emotions, using an ensemble of character- and word-level recurrent neural network models combined with a lexicon-driven system, achieving fourth place in full intensity range and third in the 0.5-1 range among 23 systems in the WASSA 2017 EmoInt shared task.
The WASSA 2017 EmoInt shared task has the goal to predict emotion intensity values of tweet messages. Given the text of a tweet and its emotion category (anger, joy, fear, and sadness), the participants were asked to build a system that assigns emotion intensity values. Emotion intensity estimation is a challenging problem given the short length of the tweets, the noisy structure of the text and the lack of annotated data. To solve this problem, we developed an ensemble of two neural models, processing input on the character. and word-level with a lexicon-driven system. The correlation scores across all four emotions are averaged to determine the bottom-line competition metric, and our system ranks place forth in full intensity range and third in 0.5-1 range of intensity among 23 systems at the time of writing (June 2017).