论文信息 - ISCLAB at SemEval-2018 Task 1: UIR-Miner for Affect in Tweets

ISCLAB at SemEval-2018 Task 1: UIR-Miner for Affect in Tweets

This paper presents a UIR-Miner system for emotion and sentiment analysis evaluation in Twitter in SemEval 2018. Our system consists of three main modules: preprocessing module, stacking module to solve the intensity prediction of emotion and sentiment, LSTM network module to solve multi-label classification, and the hierarchical attention network module for solving emotion and sentiment classification problem. According to the metrics of SemEval 2018, our system gets the final scores of 0.636, 0.531, 0.731, 0.708, and 0.408 on 5 subtasks, respectively.

[1] Saif Mohammad,et al. Understanding Emotions: A Dataset of Tweets to Study Interactions between Affect Categories , 2018, LREC.

[2] Diyi Yang,et al. Hierarchical Attention Networks for Document Classification , 2016, NAACL.

[3] Yoshua Bengio,et al. Show, Attend and Tell: Neural Image Caption Generation with Visual Attention , 2015, ICML.

[4] Jeffrey Pennington,et al. GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[5] Martin J. Wainwright,et al. Divide and Conquer Kernel Ridge Regression , 2013, COLT.

[6] Cuiying Li,et al. Partial least squares method based on least absolute shrinkage and selection operator , 2010, 2010 3rd International Conference on Advanced Computer Theory and Engineering(ICACTE).

[7] Bernhard Schölkopf,et al. A tutorial on support vector regression , 2004, Stat. Comput..

[8] J. Friedman. Greedy function approximation: A gradient boosting machine. , 2001 .

[9] Jürgen Schmidhuber,et al. Long Short-Term Memory , 1997, Neural Computation.

[10] Sankar K. Pal,et al. Multilayer perceptron, fuzzy sets, and classification , 1992, IEEE Trans. Neural Networks.

[11] T. C. Hsiang,et al. A Bayesian View on Ridge Regression , 1975 .