MITRE at SemEval-2016 Task 6: Transfer Learning for Stance Detection
Guido Zarrella,Amy Marsh +1 more
- pp 458-463
Reads0
Chats0
TLDR
MITRE's submission to the SemEval-2016 Task 6, Detecting Stance in Tweets achieved the top score in Task A on supervised stance detection, producing an average F1 score of 67.8 when assessing whether a tweet author was in favor or against a topic.Abstract:
We describe MITRE's submission to the SemEval-2016 Task 6, Detecting Stance in Tweets. This effort achieved the top score in Task A on supervised stance detection, producing an average F1 score of 67.8 when assessing whether a tweet author was in favor or against a topic. We employed a recurrent neural network initialized with features learned via distant supervision on two large unlabeled datasets. We trained embeddings of words and phrases with the word2vec skip-gram method, then used those features to learn sentence representations via a hashtag prediction auxiliary task. These sentence vectors were then fine-tuned for stance detection on several hundred labeled examples. The result was a high performing system that used transfer learning to maximize the value of the available training data.read more
Citations
More filters
Proceedings ArticleDOI
SemEval-2016 Task 6: Detecting Stance in Tweets
TL;DR: A shared task on detecting stance from tweets: given a tweet and a target entity (person, organization, etc.), automatic natural language systems must determine whether the tweeter is in favor of the given target, against thegiven target, or whether neither inference is likely.
Journal ArticleDOI
Stance and Sentiment in Tweets
TL;DR: The authors proposed a simple stance detection system that outperforms submissions from all 19 teams that participated in the SemEval-2016 shared task and showed that although knowing the sentiment expressed by a tweet is beneficial for stance classification, it alone is not sufficient.
BookDOI
A Practical Guide to Sentiment Analysis
TL;DR: The main aim of this book is to provide a feasible research platform to ambitious researchers towards developing the practical solutions that will be indeed beneficial for the authors' society, business and future researches as well.
Posted Content
Stance and Sentiment in Tweets
TL;DR: It is shown that although knowing the sentiment expressed by a tweet is beneficial for stance classification, it alone is not sufficient and additional unlabeled data is used through distant supervision techniques and word embeddings to further improve stance classification.
Posted Content
Stance Detection with Bidirectional Conditional Encoding
TL;DR: This paper used conditional LSTM encoding to detect the attitude expressed in a text towards a target such as Hillary Clinton to be "positive", negative, or neutral, and achieved state-of-the-art performance.
References
More filters
Journal ArticleDOI
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Proceedings Article
Distributed Representations of Words and Phrases and their Compositionality
TL;DR: This paper presents a simple method for finding phrases in text, and shows that learning good vector representations for millions of phrases is possible and describes a simple alternative to the hierarchical softmax called negative sampling.
Posted Content
Efficient Estimation of Word Representations in Vector Space
TL;DR: This paper proposed two novel model architectures for computing continuous vector representations of words from very large data sets, and the quality of these representations is measured in a word similarity task and the results are compared to the previously best performing techniques based on different types of neural networks.
Posted Content
How transferable are features in deep neural networks
TL;DR: This paper quantifies the generality versus specificity of neurons in each layer of a deep convolutional neural network and reports a few surprising results, including that initializing a network with transferred features from almost any number of layers can produce a boost to generalization that lingers even after fine-tuning to the target dataset.