GNN

• Graph Neural Networks for Natural Language Processing: A Survey, 2021

Term Weight

• Measuring Fine-Grained Domain Relevance of Terms, ACL2021

Knowledge Graph

• Robust Knowledge Graph Completion with Stacked Convolutions and a Student Re-Ranking Network

Information Extraction

• OntoED: Low-resource Event Detection with Ontology Embedding, ACL2021
• Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction
• PRGC: Potential Relation and Global Correspondence Based Joint Relational Triple Extraction
• PairRE: Knowledge Graph Embeddings via Paired Relation Vectors
• How Knowledge Graph and Attention Help? A Qualitative Analysis into Bag-level Relation Extraction, 2021ACL :smiley: ❤
• Fine-grained Information Extraction from Biomedical Literature based on Knowledge-enriched Abstract Meaning Representation ❤
• Dependency-driven Relation Extraction with Attentive Graph Convolutional Networks ❤
• Word Sense Disambiguation: Towards Interactive Context Exploitation from Both Word and Sense Perspectives
• PairRE: Knowledge Graph Embeddings via Paired Relation Vectors
• Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion

Text Generation

• Data Augmentation for Text Generation Without Any Augmented Data, ACL2021
• Capturing Relations between Scientific Papers: An Abstractive Model for Related Work Section Generation, 2021ACL
• Employing Argumentation Knowledge Graphs for Neural Argument Generation
• Metaphor Generation with Conceptual Mappings, 2021ACL
• AugNLG: Few-shot Natural Language Generation using Self-trained Data Augmentation, 2021ACL

NLP Machine Learning

• DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations
• Dynamic Contextualized Word Embeddings
• Unsupervised Out-of-Domain Detection via Pre-trained Transformers ❤
• Tree-Structured Topic Modeling with Nonparametric Neural Variational Inference
• Obtaining Better Static Word Embeddings Using Contextual Embedding Models
• TAN-NTM: Topic Attention Networks for Neural Topic Modeling ❤
• Learning Dense Representations of Phrases at Scale

Chat-box

• Topic-Driven and Knowledge-Aware Transformer for Dialogue Emotion Detection

多模态

• Multimodal Sentiment Detection Based on Multi-channel Graph Neural Networks

References

General Tools

• NLTK - 自然语言工具包 :+1:

• spacy - 使用 Python 和 Cython 的高性能的自然语言处理库 :+1:

• gensim - 用于对纯文本进行无监督的语义建模的库，支持 word2vec 等算法 :+1:

• StanfordNLP - 适用多语言的 NLP Library ，包含 Java 和 Python 语言 :+1:

• OpenNLP - 基于机器学习的自然语言处理的工具包，使用 Java 语言开发 :+1:

• TextBlob - 为专研常见的自然语言处理（NLP）任务提供一致的 API

• Jieba 结巴分词 - 强大的Python 中文分词库 :+1:

• HanLP - 面向生产环境的多语种自然语言处理工具包

• SnowNLP - 中文自然语言处理 Python 包，没有用NLTK，所有的算法都是自己实现的

• FudanNLP - 用于中文文本处理的 Java 函式库

• THULAC - 包括中文分词、词性标注功能。

Term Extraction

1. Bag of What Simple Noun Phrase Extraction for Text 2016. It is a pattern-based phrase extraction tool, written in Python and R.

• Basic usage of phrasemachine

  1 2 3 4 5 6

  pip install phrasemachine import phrasemachine text = "Barack Obama supports expanding social security." phrasemachine.get_phrases(text) {'num_tokens': 7, 'counts': Counter({'barack obama': 1, 'social security': 1})}

• It can support other higher accuracy spaCy tagger, or with Stanford CoreNLP.

• The position of each token can be obtained.

Ontology Query Endpoints

1. wikidata sparql 在线查询

2. SparqlEndpoints 列表 （部分不能访问）

3. 北大 gStore SPARQL Endpoint （dbpeida、freebase等）

4. http://dbpedia.org/sparql

5. Automated Phrase Mining from Massive Text Corpora 2017. This tool can be easily run by a .sh file, but needs g++, and Java as back tool.

What's the Problem?

This paper is targeted to measure the domain relevance of terms.

This blog is a water-tight version of Zhiyuanliu's tutorial on how to write a qualified NLP paper.

Process of Paper Publication

Usually, a classic process of paper publication is :

Proposal \(\longrightarrow\) Model Design \(\longrightarrow\) Coding \(\longleftrightarrow\) Parameter Tuning \(\longrightarrow\) Paper Writing \(\longleftrightarrow\) Paper Reediting \(\longrightarrow\) Paper Sumbmitting \(\longrightarrow\) Presentation

An excellet paper = An excellent work (step 1) + An excellent Writting (step 2)