Joint Learning of POS and Dependencies for Multilingual Universal Dependency Parsing (2018-01-01T00:00:00.000000Z)

TL;DR

This system predicts the part-of-speech tag and dependency tree jointly and achieves a macro-average of 68.31% LAS F1 score, with an improvement of 2.51% compared with the UDPipe.

Abstract

This paper describes the system of team LeisureX in the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Our system predicts the part-of-speech tag and dependency tree jointly. For the basic tasks, including tokenization, lemmatization and morphology prediction, we employ the official baseline model (UDPipe). To train the low-resource languages, we adopt a sampling method based on other richresource languages. Our system achieves a macro-average of 68.31% LAS F1 score, with an improvement of 2.51% compared with the UDPipe.

Authors

Zhuosheng Zhang

4 papers

Zhao Hai

3 papers

Z. Li

4 papers

TL;DR

Abstract

Authors

References36 items

CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

Seq2seq Dependency Parsing

A Full End-to-End Semantic Role Labeler, Syntactic-agnostic Over Syntactic-aware?

Deep Enhanced Representation for Implicit Discourse Relation Recognition

Syntax for Semantic Role Labeling, To Be, Or Not To Be

Subword-augmented Embedding for Cloze Reading Comprehension

Stack-Pointer Networks for Dependency Parsing

Neural Character-level Dependency Parsing for Chinese

Stack-based Multi-layer Attention for Transition-based Dependency Parsing

A Novel Neural Network Model for Joint POS Tagging and Graph-based Dependency Parsing

Adversarial Connective-exploiting Networks for Implicit Discourse Relation Classification

Deep Biaffine Attention for Neural Dependency Parsing

Graph-Based Bilingual Word Embedding for Statistical Machine Translation

Universal Dependencies v1: A Multilingual Treebank Collection

UDPipe: Trainable Pipeline for Processing CoNLL-U Files Performing Tokenization, Morphological Analysis, POS Tagging and Parsing

Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations

End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF

Converting Continuous-Space Language Models into N-gram Language Models with Efficient Bilingual Pruning for Statistical Machine Translation

Named Entity Recognition with Bidirectional LSTM-CNNs

Bilingual Continuous-Space Language Model Growing for Statistical Machine Translation

Pointer Networks

Transition-Based Dependency Parsing with Stack Long Short-Term Memory

Adam: A Method for Stochastic Optimization

Improving the Reproducibility of PAN's Shared Tasks: - Plagiarism Detection, Author Identification, and Author Profiling

Efficient Estimation of Word Representations in Vector Space

Fourth-Order Dependency Parsing

Efficient Third-Order Dependency Parsers

How Large a Corpus Do We Need: Statistical Method Versus Rule-based Method

Cross Language Dependency Parsing using a Bilingual Lexicon

Online Learning of Approximate Dependency Parsing Algorithms

Online Large-Margin Training of Dependency Parsers

Long Short-Term Memory

A Unified Syntax-aware Framework for Semantic Role Labeling

Proceedings of the 27th International Conference on Computational Linguistics

Stanford’s Graph-based Neural Dependency Parser at the CoNLL 2017 Shared Task

CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

Field of Study

Venue Information

Name

Type

URL

Alternate Names