elmargb / textmatching Goto Github PK

Understanding the Methods in Text Matching Area Including Key-words based Matching Model & Latent Semantic Matching Model. Implement the Classical Methods.

• tradition model (feature based models)
  • Key-words based methods
    • tf-idf model
    • words common rate model
    • find the most important word with adding syntax information
  • boosting models
  • linear models
  • factorization machine
• Semantic deep model
  • representation-based models
    • DSSM, CDSSM
  • interaction-based models

• Po-Sen Huang
• Jianfeng Gao
• Richard Socher
• Hang Li

深度文本匹配综述(A Survey on Deep Text Matching)

DSSM
Learning Deep Structured Semantic Models for Web Search using Clickthrough Data
CIKM 2013
词袋模型,基于语义表达的结构, word hash + DNN
详细解释
代码

CDSSM
Learning Semantic Representations Using Convolutional Neural Networks for Web Search
WWW 2014, word hash + CNN + DNN

CLSM
A Latent Semantic Model with Convolutional-Pooling Structure for Information Retrieval
CIKM 2014
基于匹配的结构, word hash + CNN, CLSM和C-DSSM有什么区别呢

DSSM的应用
Modeling Interestingness with Deep Neural Networks
EMNLP 2014
DSSM应用于文本分析，在automatic highlighting和contextual entity search问题上效果好。
主要有两点贡献：
1) DSSM + CNN
2) 不针对相关性，加了一个ranker

ARC-I/ARC-II
Convolutional Neural Network Architectures for Matching Natural Language Sentences
NIPS 2014
CNN的基于语义表达和基于匹配的两种结构; 增加了门解决句子长度不一致问题

CNTN
Convolutional Neural Tensor Network Architecture for Community-based Question Answering
IJCAI 2015
(D)CNN+MLP(tensor layer);
基于语义表达的结构

DeepMatch
A Deep Architecture for Matching Short Texts
NIPS 2013
Reviews
目的：建模更复杂的匹配关系。最早的基于匹配的结构把。
结合了localness和hierarchy intrinsic，基于点积的网络不好做的，最大的亮点是用话题模型建立网络吧。

DeepMatch_tree
Syntax-based Deep Matching of Short Texts

****

• Deep Learning for Web Search and Natural Language Processing
• Deep Learning for Information Retrieval(Sigir 2016 Tutorial)
• Semantic Matching in Search (Sigir 2014 Workshop)
• Semantic Matching in Search (Book 2014)
• gensim notebook

DSSM/Sent2Vec Release Version
MSRA发布的Sent2Vec发行版

• Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks (fb.ai/babi)
• Teaching Machines to Read and Comprehend (github.com/deepmind/rc-data)
• One Billion Word Benchmark for Measuring Progress in Statistical Language Modeling (github.com/ciprian-chelba/1-billion-word-language-modeling-benchmark)
• The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems (cs.mcgill.ca/~jpineau/datasets/ubuntu-corpus-1.0)
• Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books (BookCorpus)
• Every publicly available Reddit comment, for research.
• Stack Exchange Data Dump
• Europarl: A Parallel Corpus for Statistical Machine Translation (www.statmt.org/europarl/)
• RTE Knowledge Resources
• Kaggle Quora Question Pairs

• Kaggle Quora Question Pairs
  SQuAD(The Stanford Question Answering Dataset)

• Model Zoo
• word2vec
  • GoogleNews-vectors-negative300.bin.gz
  • freebase-vectors-skipgram1000.bin.gz
• GloVe
• SENNA

• Stanford CS224d Deep Learning for Natural Language Processing
• Stanford CS20SI Tensorflow for Deep Learning Research
• Stanford CS231n Convolutional Neural Networks for Visual Recognition

https://github.com/robertsdionne/neural-network-papers/blob/master/README.md