NTUA-SLP at SemEval-2018 Task 1: Predicting Affective Content in Tweets with Deep Attentive RNNs and Transfer Learning

TL;DR

This paper introduces a deep attentive Bi-LSTM model with transfer learning for predicting affective content in tweets, achieving top rankings in multiple subtasks of the SemEval-2018 competition.

Contribution

It presents a novel combination of self-attention, transfer learning, and affective features in a Bi-LSTM architecture for emotion prediction in tweets.

Findings

Ranked 1st in Multi-Label Emotion Classification

Ranked 2nd in Emotion Intensity Regression

Achieved competitive results in other subtasks

Abstract

In this paper we present deep-learning models that submitted to the SemEval-2018 Task~1 competition: "Affect in Tweets". We participated in all subtasks for English tweets. We propose a Bi-LSTM architecture equipped with a multi-layer self attention mechanism. The attention mechanism improves the model performance and allows us to identify salient words in tweets, as well as gain insight into the models making them more interpretable. Our model utilizes a set of word2vec word embeddings trained on a large collection of 550 million Twitter messages, augmented by a set of word affective features. Due to the limited amount of task-specific training data, we opted for a transfer learning approach by pretraining the Bi-LSTMs on the dataset of Semeval 2017, Task 4A. The proposed approach ranked 1st in Subtask E "Multi-Label Emotion Classification", 2nd in Subtask A "Emotion Intensity Regression" and achieved competitive results in other subtasks.