在学习过程中看了很多paper或教程，都存到了自己的文件夹里，但放久了自己都忘了哪篇对应哪个算法了。因此整理起来放到这里。这个列表会随着我的学习不断更新。

Pelhans/paper_list​github.com

知识图谱介绍

• 知识图谱入门笔记
• 从零开始构建知识图谱
• 徐增林_知识图谱技术综述
• 知识图谱构建技术综述

RDF 语法

• JSON-LDA JSON-based Serialization for Linked Data
• Turtle RDF 1.1 Turtle Terse RDF Triple Language
• Turtle中文翻译(自己翻译的，仅供参考)RDF 1.1 Turtle 中文翻译

结构化数据的知识抽取

从结构化数据如MYSQL等数据库获取知识得到三元组。

• 直接映射，W3C的 A Direct Mapping of Relational Data to RDF
• R2RMLR2RML: RDB to RDF Mapping Language
• D2RQ D2RQ Accessing Relational Databases as Virtual RDF Graphs

半结构化文本的知识抽取

• 面向半结构化文本的知识抽取研究
• 抽取 Web 信息的包装器归纳学习构造
• 中文百科类知识图谱的构建 http://zhishi.me Zhishi.me - Weaving Chinese Linking Open Data
• Wikipedia Mining Wikipedia as a Corpus for Knowledge Extraction
• Mining Type Information from Chinese Online Encyclopedias
• Depedia 的构建 DBpedia A Nucleus for a Web of Open Data
• DBpedia A Multilingual Cross-Domain Knowledge Base
• DBpedia - A crystallization point for the Web of Data
• DBpedia - A Large-scale, MultilingualKnowledge Base Extracted from Wikipedia

非结构化文本知识抽取

命名实体识别

• CRF++的使用
• 使用CRF++做词性标注等序列化任务
• 使用深度学习做命名实体识别Neural Architectures for Named Entity Recognition
• Bidirectional LSTM-CRF Models for Sequence Tagging
• A survey of named entity recognition and classification
• Natural language processing (almost) from scratch
• Bidirectional lstm-crf models for sequence tagging
• Neural architectures for named entity recognition
• Named entity recognition with bidirectional lstm-cnns
• Semisupervised sequence tagging with bidirectional language models
• Deep active learning for named entity recognition
• Toward mention detection robustness with recurrent neural networks
• Joint extraction of entities and relations based on a novel tagging scheme
• Fast and accurate entity recognition with iterated dilated convolutions
• Neural models for sequence chunking
• Joint extraction of multiple relations and entities by using a hybrid neural network
• End-to-end sequence labeling via bidirectional lstm-cnns-crf
• Leveraging linguistic structures for named entity recognition with bidirectional recursive neural networks
• Named entity recognition with stack residual lstm and trainable bias decoding
• Neural reranking for named entity recognition
• Deep contextualized word representations
• Attending to characters in neural sequence labeling models
• Multi-task cross-lingual sequence tagging from scratch
• Robust lexical features for improved neural network named-entity recognition
• Disease named entity recognition by combining conditional random fields and bidirectional recurrent
• Multi-channel bilstm-crf model for emerging named entity recognition in social media
• A multitask approach for named entity recognition in social media data
• Bert: Pretraining of deep bidirectional transformers for language understanding
• Named entity recognition in chinese clinical text using deep neural network
• Semi-supervised multitask learning for sequence labeling
• Efficient contextualized representation: Language model pruning for sequence labeling
• Empower sequence labeling with task-aware neural language model
• Multi-task domain adaptation for sequence tagging
• Segment-level sequence modeling using gated recursive semi-markov conditional random fields
• Hybrid semi-markov crf for neural sequence labeling
• Transfer joint embedding for crossdomain named entity recognition
• Transfer learning for sequence tagging with hierarchical recurrent networks
• Transfer learning and sentence level features for named entity recognition on tweets
• Improve neural entity recognition via multi-task data selection and constrained decoding
• Neural named entity recognition using a selfattention mechanism
• Improving clinical named entity recognition with global neural attention

关系抽取

关系抽取工具

• DeepDive 官网介绍 http://deepdive.stanford.edu/
• DeepDive 开发者毕业论文 DeepDive: A Data Management System for Automatic Knowledge Base Construction
• 支持中文的 DeepDive 并附有股权交易示例
• Deepdive实战 抽取演员-电影间关系
• 开放领域关系抽取 OpenIE 包含 TextRunner等
• TextRunner 论文

无监督方法

• 基于模板类的实体关系抽取，最简单的是基于触发词的匹配
• 复杂一点的如基于依存句法匹配的,该方法对输入的单据进行依存分析，通过依存分析输出的依存弧判断单句是否为动词谓语句，如果是则结合中文语法启发式规则抽取关系表述。根据距离确定论元位置，对三元组进行评估，输出符合条件的三元组 基于依存分析的开放式中文实体关系抽取方法
• 基于核的方法典型的为编辑距离核、字符串核、卷积树核等。基于卷积树核的方法以最短路径包含树作为关系实例的结构化表示形式，以卷积树核作为树相似度的计算方法，采用分层聚类方法进行无监督中文实体关系抽取。基于卷积树核的无指导中文实体关系抽取研究
• 基于聚类的方法，如对共现的实体及它们的上下文进行聚类，最后标记每一个类簇，以核心词汇作为关系表述。如无监督关系抽取方法研究

半监督方法

• 标签传播算法标签传播算法理论及其应用研究综述
• 标签传播算法论文Relation Extraction Using Label Propagation Based Semi-supervisedLearning
• 协同训练基于弱监督学习的海量网络数据关系抽取
• Boot Strapping 算法基于Boot Strapping的中文实体关系自动生成
• 远程监督 Distant supervision for relation extraction without labeled data

监督学习

• pipeline方法。采用Classification by Ranking CNN(CR-CNN)模型，可以学习深度学习在关系抽取中的应用方式，如position embeding 这种。Classifying Relations by Ranking with Convolutional Neural Networks
• 关系抽取中的特征介绍，包含三大类(contextual and lexical features、 nominal rol affiliation、pre-existion relation features)八小类(lexical features、hypernyms from wordNet、dependency parse、PropBank parse、FrameNet parse、nominalization、predicates from TextRunner、nominal similarity derived from the Google N-Gram data set)特征：UTD Classifying Semantic Relations by CombiningLexical and Semantic Resources
• Pipeline 方法，基于Attention CNN模型 Relation Classification via Multi-Level Attention CNNs
• Pipeline 方法，基于Attention-BLSTM模型 Attention-Based Bidirectional Long Short-Term Memory Networks forRelation Classification
• Joint 方法，基于 LSTM-RNN模型 End-to-End Relation Extraction using LSTMson Sequences and Tree Structures
• 远程监督与深度学习结合，采用注意力机制取筛选有效实例Distant Supervision for Relation Extraction with Sentence-level Attention andEntity Descriptions

事件抽取

• 对时间抽取做了一个综合的介绍，将其分为元事件抽取和主题事件抽取。其中元事件表示一个动作的发生或状态的变化。主题事件包括一类核心事件或活动以及所有与之直接相关的事件和活动，可以由多个元事件组成。对于元事件抽取包含基于模式匹配的元事件抽取和基于机器学习的元事件抽取。对于主题事件抽取，包含基于事件框架的主题事件抽取和基于本体的主题事件抽取两种。它的这个分类和我之前接触到的不大一样，仅供参考。事件抽取技术研究综述 2013 年
• 介绍神经网络在事件抽取方便的综述文章。同时对事件抽取的定义和ACE任务做了简单介绍。其中事件抽取的子任务包含触发词识别、事件类型分类、论元识别、角色分类。而触发词识别和事件类型分类又可以归结为事件识别。论元角色识别和角色分类可归为论元角色分类。根据学习方式的不同，可以将事件抽取分为基于流水线模型(先事件识别，后论元角色识别,论元角色分类的输入是上一步识别出的触发词和所有候选实体)的事件抽取和基于联合模型的事件抽取。而后，对于ACE任务，事件识别变为基于词的34类多元分类任务，角色分类变为基于词的36类多元分类任务。神经网络事件抽取技术综述
• The stages of event extraction
• Refining Event Extraction through Cross-document Inference
• Joint Event Extraction via Structured Prediction with Global Features
• Incremental Joint Extraction of Entity Mentions and Relations
• Event Extraction via Dynamic Multi-Pooling Convolutional NeuralNetworks
• Leveraging FrameNet to Improve Automatic Event Detection
• Improving Information Extraction by Acquiring External Evidence withReinforcement Learning

知识挖掘

实体消岐与链接

• Entity Linking with a Knowledge Base Issues, Techniques and Solutions
• A Generative Entity-Mention Model for Linking Entities with Knowledge Base
• Graph Ranking for Collective Named Entity Disambiguation
• Large-Scale Named Entity Disambiguation Based on Wikipedia Data
• Learning Entity Representation for Entity Disambiguation
• Learning to Link Entities with Knowledge Base
• Leveraging Deep Neural Networks and Knowledge Graphs for Entity Disambiguation
• Using TF-IDF to Determine Word Relevance in Document Queries.pdf

文本匹配

• SiameseNet – Signature Verification using a “Siamese” Time Delay Neural Network
• DSSM – Learning Deep Structured Semantic Models for Web Search using Clickthrough Data
• CDSSM – A Latent Semantic Model with Convolutional-Pooling Structure for Information Retrieval
• LSTM-DSSM – SEMANTICMODELLING WITHLONG-SHORT-TERMMEMORY FORINFORMATIONRETRIEVAL
• MV-DSSM – A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
• Learning to Rank Short Text Pairs with Convolutional Deep Neural Networks
• MatchPyramid – Text Matching as Image Recognition
• Pairwise Word Interaction Modeling with Deep Neural Networksfor Semantic Similarity Measurement
• Sentence Similarity Learning by Lexical Decomposition and Composition
• BiMPM – Bilateral Multi-Perspective Matching for Natural Language Sentences
• DecAtt – A Decomposable Attention Model for Natural Language Inference
• ESIM – Enhanced LSTM for Natural Language Inference
• A COMPARE-AGGREGATE MODEL FOR MATCHING TEXT SEQUENCES
• DAM – Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network

知识规则挖掘

• 关联规则挖掘综述
• Random walk inference and learning in a large scale knowledge base
• Variational Knowledge Graph Reasoning

知识图谱表示学习

• Collaborative Knowledge Base Embedding for Recommender Systems
• Improving Learning and Inference in a Large Knowledge-base using Latent Syntactic Cues
• Jointly Embedding Knowledge Graphs and Logical Rules
• Knowledge Graph Embedding by Translating on Hyperplanes
• Knowledge Graph Representation with Jointly Structural and Textual Encoding
• Knowledge Representation Learning with Entities, Attributes and Relations
• Learning Entity and Relation Embeddings for Knowledge Graph Completion
• Translating Embeddings for Modeling Multi-relational Data

知识存储

• 图数据库 中文第二版
• Neo4j权威指南
• 实战 将数据存进Neo4j