huangfengdan/nlp-journey

nlp相关的一些代码, 包括词向量(Word Embedding)、命名实体识别(NER)、文本分类(Text Classificatin)、文本生成、文本相似性(Text Similarity)计算等，基于keras和tensorflow

基础

• 基础知识
• 常见问题
• 实践笔记

经典书目(百度云 提取码：b5qq)

• Deep Learning.深度学习必读. 原书地址
• 斯坦福大学《语音与语言处理》第三版：NLP必读. 原书地址
• Neural Networks and Deep Learning. 入门必读. 原书地址
• 复旦大学《神经网络与深度学习》邱锡鹏教授. 原书地址
• CS224d: Deep Learning for Natural Language Processing. 课件地址

必读论文

• EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks.地址
• A Neural Probabilistic Language Model. 地址
• Transformer. 地址
• Transformer-XL. 地址
• Convolutional Neural Networks for Sentence Classification. 地址
• Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification. 地址
• A Question-Focused Multi-Factor Attention Network for Question Answering. 地址
• AutoCross: Automatic Feature Crossing for Tabular Data in Real-World Applications. 地址
• GloVe: Global Vectors for Word Representation. 官网
• A Deep Ensemble Model with Slot Alignment for Sequence-to-Sequence Natural Language Generation. 地址
• The Design and Implementation of XiaoIce, an Empathetic Social Chatbot. 地址
• A Knowledge-Grounded Neural Conversation Model. 地址
• Neural Generative Question Answering. 地址
• A Sensitivity Analysis of (and Practitioners’ Guide to) Convolutional Neural Networks for Sentence Classification. 地址
• ImageNet Classification with Deep Convolutional Neural Networks. 地址
• Network In Network. 地址,翻译
• Long Short-term Memory. 地址
• Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation. 地址
• Get To The Point: Summarization with Pointer-Generator Networks. 地址
• Generative Adversarial Text to Image Synthesis. 地址
• Image-to-Image Translation with Conditional Adversarial Networks. 地址
• Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. 地址
• Unsupervised Learning of Visual Structure using Predictive Generative Networks. 地址
• Learning to Rank Short Text Pairs with Convolutional Deep Neural Networks. 地址
• Event Extraction via Dynamic Multi-Pooling Convolutional Neural. 地址
• Low-Memory Neural Network Training:A Technical Report. 地址
• Language Models are Unsupervised Multitask Learners. 地址
• Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context. 地址
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. 地址
• SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient. 地址

必读博文

• The Illustrated Transformer.博文
• Attention-based-model. 地址
• KL divergence. 地址
• Building Autoencoders in Keras. 地址
• Modern Deep Learning Techniques Applied to Natural Language Processing. 地址
• Node2vec embeddings for graph data. 地址
• Bert解读. 地址 地址
• 难以置信！LSTM和GRU的解析从未如此清晰（动图+视频）。地址

已实现算法

• 构建词向量

• fasttext(skipgram+cbow)
• gensim(word2vec)

• 数据增强

• eda

• 分类算法

• svm
• fasttext
• textcnn
• bilstm+attention
• rcnn
• han

• NER

• bilstm+crf

• 文本相似度

• siamese

相关github项目

• keras-gpt-2. 地址
• textClassifier. 地址
• attention-is-all-you-need-keras. 地址
• BERT_with_keras. 地址
• ELMo-keras. 地址
• SeqGAN. 地址

相关博客

• 莫坠青云志
• 彗双智能-Keras源码分析
• 机器之心
• colah
• ZHPMATRIX
• wildml
• 徐阿衡
• 零基础入门深度学习

相关会议

• Association of Computational Linguistics(计算语言学协会). ACL
• Empirical Methods in Natural Language Processing. EMNLP
• International Conference on Computational Linguistics. COLING
• Neural Information Processing Systems(神经信息处理系统会议). NIPS
• AAAI Conference on Artificial Intelligence. AAAI
• International Joint Conferences on AI. IJCAI
• International Conference on Machine Learning(国际机器学习大会). ICML