Comparative Investigation of Deep Learning Components for End-to-end Implicit Discourse Relationship Parser

被引：1

作者：

Li, Dejian ^{[1
]}

Lan, Man ^{[1
]}

Wu, Yuanbin ^{[1
]}

机构：

[1] East China Normal Univ, Sch Comp Sci, Shanghai, Peoples R China

来源：

CHINESE COMPUTATIONAL LINGUISTICS, CCL 2019 | 2019年 / 11856卷

关键词：

Deep learning; Implicit discourse relation classification; Word embedding; Neural network;

D O I：

10.1007/978-3-030-32381-3_12

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The neural components in deep learning framework are crucial for the performance of many natural language processing tasks. So far there is no systematic work to investigate the influence of neural components on the performance of implicit discourse relation recognition. To address it, in this work we compare many different components and build two implicit discourse parsers base on the sequence and structure of sentence respectively. Experimental results show due to different linguistic features, the neural components have different effects in English and Chinese. Besides, our models achieve state-of-the-art performance on CoNLL-2016 English and Chinese datasets.

引用

页码：143 / 155

页数：13

共 50 条

[1] End-to-End Discourse Parser Evaluation
Ghosh, S.
Tonelli, S.
Riccardi, G.
Johansson, R.
FIFTH IEEE INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC 2011), 2011, : 169 - 172
[2] A CDT-Styled End-to-End Chinese Discourse Parser
Kong, Fang
Zhou, Guodong
ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING, 2017, 16 (04)
[3] A CDT-Styled End-to-End Chinese Discourse Parser
Kong, Fang
Wang, Hongling
Zhou, Guodong
NATURAL LANGUAGE UNDERSTANDING AND INTELLIGENT APPLICATIONS (NLPCC 2016), 2016, 10102 : 387 - 398
[4] Building a High Performance End-to-End Explicit Discourse Parser for Practical Application
Wang, Jianxiang
Lan, Man
KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2015, 2015, 9403 : 324 - 335
[5] End-to-end deep representation learning for time series clustering: a comparative study
Baptiste Lafabregue
Jonathan Weber
Pierre Gançarski
Germain Forestier
Data Mining and Knowledge Discovery, 2022, 36 : 29 - 81
[6] End-to-end deep representation learning for time series clustering: a comparative study
Lafabregue, Baptiste
Weber, Jonathan
Gancarski, Pierre
Forestier, Germain
DATA MINING AND KNOWLEDGE DISCOVERY, 2022, 36 (01) : 29 - 81
[7] End-to-End Deep Learning for Robotic Following
Pierre, John M.
ICMSCE 2018: PROCEEDINGS OF THE 2018 2ND INTERNATIONAL CONFERENCE ON MECHATRONICS SYSTEMS AND CONTROL ENGINEERING, 2015, : 77 - 85
[8] End-to-End Optimization of Deep Learning Applications
Sohrabizadeh, Atefeh
Wang, Jie
Cong, Jason
2020 ACM/SIGDA INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE GATE ARRAYS (FPGA '20), 2020, : 133 - 139
[9] Spline Filters For End-to-End Deep Learning
Balestriero, Randall
Cosentino, Romain
Glotin, Herve
Baraniuk, Richard
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 80, 2018, 80
[10] End-to-end deep learning with neuromorphic photonics
Dabos, G.
Mourgias-Alexandris, G.
Totovic, A.
Kirtas, M.
Passalis, N.
Tefas, A.
Pleros, N.
INTEGRATED OPTICS: DEVICES, MATERIALS, AND TECHNOLOGIES XXV, 2021, 11689

← 1 2 3 4 5 →