DataStories at SemEval-2017 Task 4: Deep LSTM with Attention for Message-level and Topic-based Sentiment Analysis
By Christos Baziotis, Nikos Pelekis, Christos Doulkeridis. International Workshop on Semantic Evaluation, 2017
484 citations recorded by the source corpus.
In this paper we present two deep-learning systems that competed at SemEval-2017 Task 4 “Sentiment Analysis in Twitter”. We participated in all subtasks for English tweets, involving message-level and topic-based sentiment polarity classification and quantification. We use Long Short-Term Memory (LSTM) networks augmented with two kinds of attention mechanisms, on top of word embeddings pre-trained on a big collection of Twitter messages. Also, we present a text processing tool suitable for social network messages, which performs tokenization, word normalization, segmentation and spell correction. Moreover, our approach uses no hand-crafted features or sentiment lexicons. We ranked 1st (tie) in Subtask A, and achieved very competitive results in the rest of the Subtasks. Both the word embeddings and our text processing tool are available to the research community.
Subfields: Artificial Intelligence, Management Science and Operations Research
View Paper: https://www.semanticscholar.org/paper/3b2ea67e3fa3182808c8c5f4a7edff354fb16cb6
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Guangxuan Xu
- Alexandros Potamianos
- Soroush Vosoughi
- Ruibo Liu
- Henry Nunoo-Mensah
- Francisca Adoma Acheampong
- Wenyu Chen
- Alexandra Chronopoulou
In Management Science and Operations Research
- Henry Nunoo-Mensah
- Francisca Adoma Acheampong
- Alexandros Potamianos
- Wenyu Chen
- Alexandra Chronopoulou
- Christos Baziotis
Citations in the corpus
- Data Boost: Text Data Augmentation through Reinforcement Learning Guided Conditional Generation (2020, background): “Sentiment Analysis, though seemingly well-resolved according to some literature (Baziotis et al., 2017; Cliche, 2017), is reported to have severe overfitting problems when given extremely limited training data (Elming et al., 2014; Severyn and Moschitti, 2015).”
- An Embarrassingly Simple Approach for Transfer Learning from Pretrained Language Models (2019, methodology): “To preprocess the tweets, we use Ekphrasis (Baziotis et al., 2017).”
- SemEval-2019 Task 3: EmoContext Contextual Emotion Detection in Text (2019, methodology): “A good number of teams used the “ekphrasis” package (Baziotis et al., 2017) for tokenization, word normalization and word segmentation.”
- Transformer models for text-based emotion detection: a review of BERT-based approaches (2021, methodology): “The authors used the ekphrasis package (Baziotis et al. 2017) to convert the emojis to their textual aliases and further cleaned the converted texts.”
- Sentiment Analysis of Comment Texts Based on BiLSTM (2019, unknown): “[22] introduced the attention mechanism into the LSTM, which achieved good results in the sentiment analysis of SemEval-2017 Task4 for Twitter.”