A Multi-Task Neural Network for Multilingual Sentiment Classification and Language Detection on Twitter

被引：16

作者：

Wehrmann, Jonatas ^{[1
]}

Becker, Willian E. ^{[1
]}

Barros, Rodrigo C. ^{[1
]}

机构：

[1] Pontificia Univ Catolica Rio Grande do Sul, Porto Alegre, RS, Brazil

来源：

33RD ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING | 2018年

关键词：

multitask classification; sentiment analysis; language detection; convolutional neural networks;

D O I：

10.1145/3167132.3167325

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this paper, we propose a novel approach for classifying both the sentiment and the language of tweets. Our proposed architecture comprises a convolutional neural network (ConvNet) with two distinct outputs, each of which designed to minimize the classification error of either sentiment assignment or language identification. Results show that our method outperforms both single-task and multi-task state-of-the-art approaches for classifying multilingual tweets.

引用

页码：1805 / 1812

页数：8

共 39 条

[1]

Alexandru-Lucian G^insca, 2011, Proceedings of the 2nd Workshop on Computational Approaches to Sub- jectivity and Sentiment Analysis, P189

[2]

[Anonymous], 2004, Proceedings of the 42nd annual meeting on Association for Computational Linguistics, DOI DOI 10.3115/1218955.1218990

[3]

[Anonymous], 2017, APPL SOFT COMPUTING

[4]

[Anonymous], 2012, P ACL 2012 SYST DEM, DOI 10.1145/1935826.1935854

[5]

[Anonymous], AUTOMATIC IDENTIFICA

[6]

[Anonymous], 2007, P INEWS07 WORKSH SIG

[7]

[Anonymous], PATTERN RECOGNITION

[8]

[Anonymous], 2015, NIPS

[9]

[Anonymous], 2016, ARXIV161003017

[10]

[Anonymous], 2014, C EMPIRICAL METHODS

← 1 2 3 4 →