天然語言處理(NLP)是計算機科學,人工智能,語言學關注計算機和人類(天然)語言之間的相互做用的領域。本文做者爲天然語言處理NLP初學者整理了一份龐大的天然語言處理項目領域的概覽,包括了不少人工智能應用程序。選取的參考文獻與資料都側重於最新的深度學習研究成果。這些天然語言處理項目資源能爲想要深刻鑽研一個天然語言處理NLP任務的人們提供一個良好的開端。php
https://github.com/Kyubyong/nlp_tasks#coreference-resolutionhtml
- 論文:Automatic Text Scoring Using Neural Networks(使用神經網絡的自動文本評分):https://arxiv.org/abs/1606.04289
- 論文:A Neural Approach to Automated Essay Scoring(一種自動將論文評分的神經學方法):http://www.aclweb.org/old_anthology/D/D16/D16-1193.pdf
- 挑戰:Kaggle:The Hewlett Foundation: Automated Essay Scoring(Kaggle:The Hewlett
Foundation:論文自動評分系統):https://www.kaggle.com/c/asap-aes- 項目:Enhanced AI Scoring Engine(加強的人工智能得分引擎):https://github.com/edx/ease
- 維基百科: 語言識別:https://en.wikipedia.org/wiki/Speech_recognition
- 論文:DeepSpeech 2: End-to-End Speech Recognition in English and Mandarin(深度語音2:用英語和普通話進行端對端語音識別):https://arxiv.org/abs/1512.02595
- 論文:WaveNet:A Generative Model for Raw Audio(WaveNet:原始音頻的生成模型):https://arxiv.org/abs/1609.03499
- 項目:A TensorFlow implementation of Baidu’s Deep Speech architecture(百度深度語音架構的一個TensorFlow實現:https://github.com/mozilla/DeepSpeech
- 項目:Speech-to-Text-WaveNet: End-to-end sentence level English speech recognition using DeepMind’s WaveNet(Speech-to-Text-WaveNet:
使用DeepMind的WaveNet,對端到端句子的英語水平語音識別):https://github.com/buriburisuri/speech-to-text-wavenet- 挑戰:The 5th CHiME Speech Separation and Recognition Challenge(第五屆CHiME語音的分離和識別挑戰):http://spandh.dcs.shef.ac.uk/chime_challenge/
- 資料:The 5thCHiME Speech Separation and Recognition Challenge(第五屆CHiME語音的分離和識別挑戰):http://spandh.dcs.shef.ac.uk/chime_challenge/download.html
- 資料:CSTRVCTK Corpus :http://homepages.inf.ed.ac.uk/jyamagis/page3/page58/page58.html
- 資料:LibriSpeech ASR corpus:http://www.openslr.org/12/
- 資料:Switchboard-1 Telephone Speech Corpus:https://catalog.ldc.upenn.edu/ldc97s62
- 資料:TED-LIUM Corpus:http://www-lium.univ-lemans.fr/en/content/ted-lium-corpus
- 維基百科:自動摘要:https://en.wikipedia.org/wiki/Automatic_summarization
- 書籍:Automatic Text Summarization(自動本文摘要):https://www.amazon.com/Automatic-Text-Summarization-Juan-Manuel-Torres-Moreno/dp/1848216688/ref=sr_1_1?s=books&ie=UTF8&qid=1507782304&sr=1-1&keywords=Automatic+Text+Summarization
- 論文:Text Summarization Using Neural Networks(使用神經網絡進行文本摘要):http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.823.8025&rep=rep1&type=pdf
- 論文:Ranking with Recursive Neural Networks and Its Application to Multi-DocumentSummarization(使用遞歸神經網絡及其應用程序對多文檔摘要進行排序):https://www.aaai.org/ocs/index.php/AAAI/AAAI15/paper/viewFile/9414/9520
- 資料:Text Analytics Conferences(文本分析會議):https://tac.nist.gov/data/index.html
- 資料:Document Understanding Conferences(文書理解會議):http://www-nlpir.nist.gov/projects/duc/data.html
- 信息:共指消解:https://nlp.stanford.edu/projects/coref.shtml
- 論文:Deep Reinforcement Learning for Mention-Ranking Coreference Models(對Mention-Ranking的共指模型進行深度強化學習:https://arxiv.org/abs/1609.08667
- 論文:Improving Coreference Resolution by Learning Entity-Level Distributed
Representations(經過學習實體級分佈式表示來改善相關的解決方案):https://arxiv.org/abs/1606.01323- 挑戰:CoNLL 2012 Shared Task: Modeling Multilingual Unrestricted Coreference in OntoNotes(CoNLL
2012共享任務:在OntoNotes中對多語言的不受限制的共指進行建模):http://conll.cemantix.org/2012/task-description.html- 挑戰:CoNLL 2011 Shared Task: Modeling Unrestricted Coreference in OntoNotes(CoNLL
2011共享任務:在OntoNotes中對多語言的不受限制的共指進行建模):http://conll.cemantix.org/2011/task-description.html
- 論文:Neural Network Translation Models for Grammatical Error Correction(語法錯誤校訂的神經網絡翻譯模型):https://arxiv.org/abs/1606.00189
- 挑戰:CoNLL 2013 Shared Task: Grammatical Error Correction(CoNLL 2013共享任務:語法錯誤校訂):http://www.comp.nus.edu.sg/~nlp/conll13st.html
- 挑戰:CoNLL 2014Shared Task: Grammatical Error Correction(CoNLL 2014共享任務:語法錯誤校訂):http://www.comp.nus.edu.sg/~nlp/conll14st.html
- 資料:NUSNon-commercial research/trial corpus license:http://www.comp.nus.edu.sg/~nlp/conll14st/nucle_license.pdf
- 資料:Lang-8 Learner Corpora:http://cl.naist.jp/nldata/lang-8/
- 資料:Cornell Movie–Dialogs Corpus:http://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html
- 項目:Deep Text Corrector(深度文本校訂器):https://github.com/atpaino/deep-text-corrector
- 產品:deep grammar:http://deepgrammar.com/
- 論文:Grapheme-to-Phoneme Models for (Almost)Any Language(適合(幾乎)任何語言的字素到音素的模型):
https://pdfs.semanticscholar.org/b9c8/fef9b6f16b92c6859f6106524fdb053e9577.pdf- 論文:Polyglot Neural Language Models: A Case Study in Cross-Lingual Phonetic Representation
Learning(多語言神經語言模型:跨語語音表達學習的案例研究):
https://arxiv.org/pdf/1605.03832.pdf- 論文:Multi task Sequence-to-Sequence Models for Grapheme-to-Phoneme Conversion(多任務序列到序列的字素到音素轉換的模型):
https://pdfs.semanticscholar.org/26d0/09959fa2b2e18cddb5783493738a1c1ede2f.pdf- 項目:Sequence-to-Sequence G2P toolkit(序列到序列G2P工具包):
https://github.com/cmusphinx/g2p-seq2seq- 資料:Multilingual Pronunciation Data(多語種發音數據):
https://drive.google.com/drive/folders/0B7R_gATfZJ2aWkpSWHpXUklWUmM
- 維基百科: 語種識別:https://en.wikipedia.org/wiki/Language_identification
- 論文:AUTOMATIC LANGUAGE IDENTIFICATION USING DEEP NEURAL NETWORKS(使用深度神經網絡的自動語言識別):
https://repositorio.uam.es/bitstream/handle/10486/666848/automatic_lopez-moreno_ICASSP_2014_ps.pdf?sequence=1- 挑戰: 2015 Language Recognition Evaluation(2015語言識別評估):
https://www.nist.gov/itl/iad/mig/2015-language-recognition-evaluation
- 維基百科:語言模型:https://en.wikipedia.org/wiki/Language_model
- 工具包: KenLM Language Model Toolkit(KenLM語言模型工具包):
http://kheafield.com/code/kenlm/- 論文:Distributed Representations of Words and Phrases and their Compositionality(詞彙和短語的分佈表示及其組合性):
http://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf- 論文:Character-Aware Neural Language Models(Character-Aware神經語言模型):
https://www.aaai.org/ocs/index.php/AAAI/AAAI16/paper/viewFile/12489/12017- 資料: Penn Treebank :
https://github.com/townie/PTB-dataset-from-Tomas-Mikolov-s-webpage/tree/master/data
- 維基百科:詞形還原:https://en.wikipedia.org/wiki/Lemmatisation
- 工具包:WordNet Lemmatizer:
http://www.nltk.org/api/nltk.stem.html#nltk.stem.wordnet.WordNetLemmatizer.lemmatize- 資料:Treebank-3:https://catalog.ldc.upenn.edu/ldc99t42
- 維基百科:脣讀法:https://en.wikipedia.org/wiki/Lip_reading
- 論文:Lip Reading Sentences in the Wild (在野外讀懂脣語):
https://arxiv.org/abs/1611.05358
https://arxiv.org/abs/1706.05739- 項目: Lip Reading – Cross Audio-Visual Recognition using 3D Convolutional Neural
Networks(脣讀法—使用3D卷積神經網絡的交叉視聽識別:
https://github.com/astorfi/lip-reading-deeplearning- 資料: The GRID audiovisual sentence corpus:
http://spandh.dcs.shef.ac.uk/gridcorpus/
- 論文:Neural Machine Translation by Jointly Learning to Align and Translate(經過共同窗習來調整和翻譯神經機器翻譯):
https://arxiv.org/abs/1409.0473- 論文:Neural Machine Translation in Linear Tim(在線性時間中的神經機器翻譯):
https://arxiv.org/abs/1610.10099- 挑戰: ACL2014 NINTH WORKSHOP ON STATISTICAL MACHINE TRANSLATION(ACL2014第九屆統計機器翻譯研討會):
http://www.statmt.org/wmt14/translation-task.html#download- 資料:OpenSubtitles2016:http://opus.lingfil.uu.se/OpenSubtitles2016.php
- 資料: WIT3:Web Inventory of Transcribed and Translated Talks:https://wit3.fbk.eu/
- 資料: The QCRI Educational Domain (QED) Corpus:
http://alt.qcri.org/resources/qedcorpus/
- 維基百科:命名實體識別:https://en.wikipedia.org/wiki/Named-entity_recognition
- 論文:Neural Architectures for Named Entity Recognition(命名實體識別的神經結構):https://arxiv.org/abs/1603.01360
- 項目: OSU Twitter NLP Tool:https://github.com/aritter/twitter_nlp
- 挑戰: Named Entity Recognition in Twitter(在推特上被命名的實體識別):
https://noisy-text.github.io/2016/ner-shared-task.html- 資料:CoNLL-2002 NER corpus:
https://github.com/teropa/nlp/tree/master/resources/corpora/conll2002- 資料:CoNLL-2003 NER corpus:
https://github.com/synalp/NER/tree/master/corpus/CoNLL-2003
- 論文:Dynamic Pooling and Unfolding Recursive Autoencoders for Paraphrase
Detection(動態池和展開遞歸自動編碼器的釋義檢測):
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.650.7199&rep=rep1&type=pdf- 項目:Paralex: Paraphrase-Driven Learning for Open Question Answering(Paralex:釋義驅動學習的開放問答):
http://knowitall.cs.washington.edu/paralex/- 資料:Microsoft Research Paraphrase Corpus:
https://www.microsoft.com/en-us/download/details.aspx?id=52398- 資料:Microsoft Research Video Description Corpus :
https://www.microsoft.com/en-us/download/details.aspx?id=52422&from=http%3A%2F%2Fresearch.microsoft.com%2Fen-us%2Fdownloads%2F38cf15fd-b8df-477e-a4e4-a4680caa75af%2F- 資料: Pascal Dataset:
http://nlp.cs.illinois.edu/HockenmaierGroup/pascal-sentences/index.html- 資料:Flicker Dataset:http://nlp.cs.illinois.edu/HockenmaierGroup/8k-pictures.html
- 資料: TheSICK data set:http://clic.cimec.unitn.it/composes/sick.html
- 資料: PPDB:The Paraphrase Database:http://www.cis.upenn.edu/~ccb/ppdb/
- 資料:WikiAnswers Paraphrase Corpus:
http://knowitall.cs.washington.edu/paralex/wikianswers-paraphrases-1.0.tar.gz
- 維基百科:語法分析:https://en.wikipedia.org/wiki/Parsing
- 工具包:The Stanford Parser: A statistical parser:https://nlp.stanford.edu/software/lex-parser.shtml
- 工具包: spaCyparser:https://spacy.io/docs/usage/dependency-parse
- 論文:A fastand accurate dependency parser using neural networks(快速而準確地使用神經網絡的依賴解析器):http://www.aclweb.org/anthology/D14-1082
- 挑戰:CoNLL2017 Shared Task: Multilingual Parsing from Raw Text to Universal
Dependencies(CoNLL2017共享任務:從原始文本到通用依賴項的多語言解析):http://universaldependencies.org/conll17/- 挑戰:CoNLL2016 Shared Task: Multilingual Shallow Discourse Parsing(CoNLL2016共享任務:多語言的淺會話解析):http://www.cs.brandeis.edu/~clp/conll16st/
- 維基百科:詞性標記:https://en.wikipedia.org/wiki/Part-of-speech_tagging
- 論文:Unsupervised Part-Of-Speech Tagging with Anchor Hidden Markov Models(有Anchor Hidden
Markov模型的非監督性的詞性標記)
https://transacl.org/ojs/index.php/tacl/article/viewFile/837/192- 資料:Treebank-3:https://catalog.ldc.upenn.edu/ldc99t42
- 工具包:nltk.tag package:http://www.nltk.org/api/nltk.tag.html
- 論文:Neural Network Language Model for Chinese Pinyin Input Method Engine(中文拼音輸入法引擎的神經網絡語言模型):
http://aclweb.org/anthology/Y15-1052- 項目:Neural Chinese Transliterator:
https://github.com/Kyubyong/neural_chinese_transliterator
- 維基百科:問答系統:https://en.wikipedia.org/wiki/Question_answering
- 論文:Ask Me Anything: Dynamic Memory Networks for Natural Language Processing(天然語言處理的動態內存網絡):
http://www.thespermwhale.com/jaseweston/ram/papers/paper_21.pdf- 論文:Dynamic Memory Networks for Visual and Textual Question Answering(用於視覺和文本的問答系統的動態記憶網絡):
http://proceedings.mlr.press/v48/xiong16.pdf- 挑戰:TREC Question Answering Task(TREC問答系統任務):
http://trec.nist.gov/data/qamain.html- 挑戰:SemEval-2017 Task 3: Community Question Answering:
http://alt.qcri.org/semeval2017/task3/- 資料:MSMARCO: Microsoft MAchine Reading COmprehension Dataset(MSMARCO:微軟機器閱讀理解數據集)http://www.msmarco.org/
- 資料:Maluuba NewsQA:https://github.com/Maluuba/newsqa
- 資料:SQuAD:100,000+ Questions for Machine Comprehension of Text(SQuAD:100,000+個文本的機器理解的問題):https://rajpurkar.github.io/SQuAD-explorer/
- 資料:Graph Questions: A Characteristic-rich Question Answering Dataset(圖形問題:一個特徵豐富的問題回答數據集):https://github.com/ysu1989/GraphQuestions
- 資料: Story Cloze Test and ROC Stories Corpora:
http://cs.rochester.edu/nlp/rocstories/- 資料:Microsoft Research WikiQA Corpus:
https://www.microsoft.com/en-us/download/details.aspx?id=52419&from=http%3A%2F%2Fresearch.microsoft.com%2Fen-us%2Fdownloads%2F4495da01-db8c-4041-a7f6-7984a4f6a905%2Fdefault.aspx- 資料:DeepMind Q&A Dataset:http://cs.nyu.edu/~kcho/DMQA/
- 資料: QASent:http://cs.stanford.edu/people/mengqiu/data/qg-emnlp07-data.tgz
- 維基百科:關係提取:https://en.wikipedia.org/wiki/Relationship_extraction
- 論文:A deep learning approach for relationship extraction from interaction context in social manufacturing
paradigm(一種從社會生產範例的互動情境中提取關係深度學習的方法):http://www.sciencedirect.com/science/article/pii/S0950705116001210
- 維基百科:語義角色標記:https://en.wikipedia.org/wiki/Semantic_role_labeling
- 書籍:Semantic Role Labeling(語義角色標記):
https://www.amazon.com/Semantic-Labeling-Synthesis-Lectures-Technologies/dp/1598298313/ref=sr_1_1?s=books&ie=UTF8&qid=1507776173&sr=1-1&keywords=Semantic+Role+Labeling- 論文:End-to-end Learning of Semantic Role Labeling Using Recurrent Neural Networks(使用循環神經網絡對語義角色標籤進行端到端學習):
http://www.aclweb.org/anthology/P/P15/P15-1109.pdf- 論文:Neural Semantic Role Labeling with Dependency Path Embeddings(有着依賴路徑嵌入的神經語義角色標記):
https://arxiv.org/abs/1605.07515
挑戰:CoNLL-2005 Shared Task: Semantic Role
Labeling(CoNLL-2005共享任務:語義角色標記)
http://www.cs.upc.edu/~srlconll/st05/st05.html- 挑戰:CoNLL-2004 Shared Task: Semantic Role Labeling(CoNLL-2004共享任務:語義角色標記):
http://www.cs.upc.edu/~srlconll/st04/st04.html- 工具包:Illinois Semantic Role Labeler(SRL):http://cogcomp.org/page/software_view/SRL
- 資料:CoNLL-2005 Shared Task: Semantic Role Labeling(CoNLL-2005共享任務:語義角色標記):
http://www.cs.upc.edu/~srlconll/soft.html
- 維基百科:語句邊界消歧:https://en.wikipedia.org/wiki/Sentence_boundary_disambiguation
- 論文:A Quantitative and Qualitative Evaluation of Sentence Boundary Detection for theClinical Domain(對臨牀領域的語句邊界檢測進行定量和定性的評估):
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5001746/- 工具包: NLTK Tokenizers:http://www.nltk.org/_modules/nltk/tokenize.html
- 資料: The British National Corpus:http://www.natcorp.ox.ac.uk/
- 資料:Switchboard-1 Telephone Speech Corpus:https://catalog.ldc.upenn.edu/ldc97s62
- 維基百科:情緒分析:https://en.wikipedia.org/wiki/Sentiment_analysis
- 信息:Awesome Sentiment Analysis(了不得的情緒分析):
https://github.com/xiamx/awesome-sentiment-analysis- 挑戰:Kaggle: UMICH SI650 – Sentiment Classification(Kaggle: UMICH SI650 – 情緒分類):
https://www.kaggle.com/c/si650winter11#description- 挑戰:SemEval-2017 Task 4: Sentiment Analysis in Twitter(SemEval-2017任務4:推特上的情緒分析):
http://alt.qcri.org/semeval2017/task4/- 項目:SenticNet:http://sentic.net/about/
- 資料:Multi-Domain Sentiment Dataset(version2.0):
http://www.cs.jhu.edu/~mdredze/datasets/sentiment/- 資料:Stanford Sentiment Treebank:https://nlp.stanford.edu/sentiment/code.html
- 資料:Twitter Sentiment Corpus:http://www.sananalytics.com/lab/twitter-sentiment/
- 資料:Twitter Sentiment Analysis Training Corpus:http://thinknook.com/twitter-sentiment-analysis-training-corpus-dataset-2012-09-22/
- 維基百科:源分離:https://en.wikipedia.org/wiki/Source_separation
- 論文:From Blind to Guided Audio Source Separation(從盲目到有指導性的音頻源分離):https://hal-univ-rennes1.archives-ouvertes.fr/hal-00922378/document
- 論文:Joint Optimization of Masks and Deep Recurrent Neural Networks for Monaural Source Separation
(對單聲道分離的掩膜和深層循環神經網絡的聯合優化):https://arxiv.org/abs/1502.04149- 挑戰:Signal Separation Evaluation Campaign(信號分離評估活動):https://sisec.inria.fr/
- 挑戰: CHiME Speech Separation and Recognition Challenge(CHiME語音分離和識別的挑戰):http://spandh.dcs.shef.ac.uk/chime_challenge/
- 維基百科:說話者識別:https://en.wikipedia.org/wiki/Speaker_recognition
- 論文:A NOVEL SCHEME FOR SPEAKER RECOGNITION USING A PHONETICALLY-AWARE DEEP NEURAL
NETWORK(一種使用語音識別的深度神經網絡的新方案):
https://pdfs.semanticscholar.org/204a/ff8e21791c0a4113a3f75d0e6424a003c321.pdf- 論文:DEEP NEURAL NETWORKS FOR SMALL FOOTPRINT TEXT-DEPENDENT SPEAKER VERIFICATION(深度神經網絡,用於小範圍的文本依賴的說話者驗證):https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/41939.pdf
- 挑戰: NIST Speaker Recognition Evaluation(NIST說話者識別評價):https://www.nist.gov/itl/iad/mig/speaker-recognition
- 維基百科:語音分段:https://en.wikipedia.org/wiki/Speech_segmentation
- 論文:Word Segmentation by 8-Month-Olds: When Speech Cues Count More Than
Statistics(8個月大嬰兒的單詞分段:當語音提示比統計數字更重要時):http://www.utm.toronto.edu/infant-child-centre/sites/files/infant-child-centre/public/shared/elizabeth-johnson/Johnson_Jusczyk.pdf- 論文:Unsupervised Word Segmentation and Lexicon Discovery Using Acoustic Word
Embeddings(不受監督的單詞分割和使用聲學詞嵌入的詞彙發現):https://arxiv.org/abs/1603.02845- 資料:CALLHOME Spanish Speech:https://catalog.ldc.upenn.edu/ldc96s35
- 維基百科:語音合成:https://en.wikipedia.org/wiki/Speech_synthesis
- 論文:WaveNet:A Generative Model for Raw Audio(WaveNet:原始音頻的生成模型):https://arxiv.org/abs/1609.03499
- 論文:Tacotron:Towards End-to-End Speech Synthesis(Tacotron:對端到端的語音合成):https://arxiv.org/abs/1703.10135
- 資料: The World English Bible:https://github.com/Kyubyong/tacotron
- 資料: LJ Speech Dataset:https://github.com/keithito/tacotron
- 資料: Lessac Data:http://www.cstr.ed.ac.uk/projects/blizzard/2011/lessac_blizzard2011/
- 挑戰:Blizzard Challenge 2017:https://synsig.org/index.php/Blizzard_Challenge_2017
- 項目: The Festvox project:http://www.festvox.org/index.html
- 工具包:Merlin: The Neural Network (NN) based Speech Synthesis System(Merlin:基於神經網絡的語音合成系統):https://github.com/CSTR-Edinburgh/merlin
- 維基百科:語音加強:https://en.wikipedia.org/wiki/Speech_enhancement
- 書籍: Speech enhancement: theory and practice(語音加強:理論與實踐):
https://www.amazon.com/Speech-Enhancement-Theory-Practice-Second/dp/1466504218/ref=sr_1_1?ie=UTF8&qid=1507874199&sr=8-1&keywords=Speech+enhancement%3A+theory+and+practice- 論文 An Experimental Study on Speech Enhancement Based on Deep Neural Network(一項基於深度神經網絡的語音加強實驗):
http://staff.ustc.edu.cn/~jundu/Speech%20signal%20processing/publications/SPL2014_Xu.pdf- 論文: A Regression Approach to Speech Enhancement Based on Deep Neural
https://www.researchgate.net/profile/Yong_Xu63/publication/272436458_A_Regression_Approach_to_Speech_Enhancement_Based_on_Deep_Neural_Networks/links/57fdfdda08aeaf819a5bdd97.pdf- 論文:Speech Enhancement Based on Deep Denoising Autoencoder(基於深度降噪自編碼的語音加強):
https://www.researchgate.net/profile/Yu_Tsao/publication/283600839_Speech_enhancement_based_on_deep_denoising_Auto-Encoder/links/577b486108ae213761c9c7f8/Speech-enhancement-based-on-deep-denoising-Auto-Encoder.pdf
- 維基百科:詞幹提取:https://en.wikipedia.org/wiki/Stemming
- 論文: A BACKPROPAGATION NEURAL NETWORK TO IMPROVE ARABIC STEMMING(一個反向傳播的神經網絡,用來改善阿拉伯語的詞幹提取):
http://www.jatit.org/volumes/Vol82No3/7Vol82No3.pdf- 工具包: NLTK Stemmers:http://www.nltk.org/howto/stem.html
- 維基百科:術語提取:https://en.wikipedia.org/wiki/Terminology_extraction
- 論文: Neural Attention Models for Sequence Classification: Analysis and Application to KeyTerm Extraction and Dialogue Act
Detection(序列分類的神經提示模型:分析和應用於關鍵詞提取和對話法檢測):https://arxiv.org/pdf/1604.00077.pdf
- 維基百科:文本簡化:https://en.wikipedia.org/wiki/Text_simplification
- 論文:Aligning Sentences from Standard Wikipedia to Simple Wikipedia(調整句子,從標準的維基百科到簡單的維基百科):
https://ssli.ee.washington.edu/~hannaneh/papers/simplification.pdf- 論文:Problems in Current Text Simplification Research: New Data Can Help(當前文本簡化研究中的問題:可提供幫助的新數據):
https://pdfs.semanticscholar.org/2b8d/a013966c0c5e020ebc842d49d8ed166c8783.pdf- 資料:Newsela Data:https://newsela.com/data/
- 維基百科:文本蘊含:https://en.wikipedia.org/wiki/Textual_entailment
- 項目:Textual Entailment with TensorFlow(文本蘊含與TensorFlow):
https://github.com/Steven-Hewitt/Entailment-with-Tensorflow- 競賽:SemEval-2013 Task 7: The Joint Student Response Analysis and 8th Recognizing Textual Entailment Challenge(SemEval-2013任務7:聯合學生反應分析和第8屆認知文本蘊含挑戰):
https://www.cs.york.ac.uk/semeval-2013/task7.html
- 維基百科:音譯:https://en.wikipedia.org/wiki/Transliteration
- 論文:A Deep Learning Approach to Machine Transliteration(一個機器音譯的深度學習方法):https://pdfs.semanticscholar.org/54f1/23122b8dd1f1d3067cf348cfea1276914377.pdf
- 項目:Neural Japanese Transliteration—can you do better than SwiftKey™ Keyboard?(神經日語音譯:你能比SwiftKey鍵盤作得更好嗎?):
https://github.com/Kyubyong/neural_japanese_transliterator
- 維基百科:詞嵌入:https://en.wikipedia.org/wiki/Word_embedding
- 工具包:Gensim: word2vec:https://radimrehurek.com/gensim/models/word2vec.html
- 工具包:fastText:https://github.com/facebookresearch/fastText
- 工具包:GloVe:Global Vectors for Word Representation:
https://nlp.stanford.edu/projects/glove/- 信息:Where to get a pretrained model?(哪裏可以得到一個預先訓練的模型?):https://github.com/3Top/word2vec-api
- 項目:Pre-trained word vectors of 30+ languages(30多種語言的預先訓練的詞向量):https://github.com/Kyubyong/wordvectors
- 項目:Polyglot: Distributed word representations for multilingual NLP(Polyglot:多語言NLP的分佈式詞彙表徵):https://sites.google.com/site/rmyeid/projects/polyglot
- 信息:What is Word Prediction?(什麼是詞彙預測?):
http://www2.edc.org/ncip/library/wp/what_is.htm- 論文: The prediction of character based on recurrent neural network language
model(基於循環神經網絡語言模型的字符預測):http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7960065- 論文: An Embedded Deep Learning based Word Prediction(一個基於深度學習的詞彙預測):https://arxiv.org/abs/1707.01662
- 論文:Evaluating Word Prediction: Framing Keystroke Savings(評估單詞預測:框擊鍵保存):http://aclweb.org/anthology/P08-2066
- 資料:An Embedded Deep Learning based Word Prediction(一個基於深度學習的詞彙預測):https://github.com/Meinwerk/WordPrediction/master.zip
- 項目: Word Prediction using Convolutional Neural Networks—can you do better than iPhone™
Keyboard?(使用卷積神經網絡的詞彙預測——你能比iPhone鍵盤作得更好嗎?):https://github.com/Kyubyong/word_prediction
- 論文: Neural Word Segmentation Learning for Chinese(中文的神經詞分割學習):https://arxiv.org/abs/1606.04300
- 項目:Convolutional neural network for Chinese word segmentation(中文的詞分割的卷積神經網絡):https://github.com/chqiwang/convseg
- 工具包:Stanford Word Segmenter:https://nlp.stanford.edu/software/segmenter.html
- 工具包: NLTK Tokenizers:http://www.nltk.org/_modules/nltk/tokenize.html
- 維基百科:詞義消歧:https://en.wikipedia.org/wiki/Word-sense_disambiguation
- 論文:Train-O-Matic: Large-Scale Supervised Word Sense Disambiguation in Multiple Languages without Manual Training
Data(Train-O-Matic:在沒有人工訓練數據的狀況下,在多 -
種語言中大規模的監督詞義消歧):http://www.aclweb.org/anthology/D17-1008- 資料:Train-O-Matic Data:http://trainomatic.org/data/train-o-matic-data.zip
- 資料:BabelNet:http://babelnet.org/
- 原項目地址:https://github.com/Kyubyong/nlp_tasks#speech-segmentation
karpathy/char-rnn · GitHub :一個基於RNN的文本生成器。能夠自動生成莎士比亞的劇本或者shell代碼。
https://github.com/karpathy/char-rnngit
phunterlau/wangfeng-rnn · GitHub : 基於char-rnn的汪峯歌詞生成器
https://github.com/phunterlau/wangfeng-rnngithub
google/deepdream · GitHub :畫出神經網絡眼中的世界
https://github.com/google/deepdreamweb
facebook/MemNN · GitHub :memnn的一個官方實現。能夠回答諸如「小明在操場;小王在辦公室;小明撿起了足球;小王走進了廚房。問:小王在去廚房前在哪裏?」,這樣涉及推理和理解的問題。
https://github.com/facebook/MemNNshell
skaae/lasagne-draw · GitHub :用RNN生成手寫數字。
https://github.com/skaae/lasagne-drawapi
keras/addition_rnn.py at master · fchollet/keras · GitHub :用RNN自動學會加法規則。
https://github.com/keras-team/keras/blob/master/examples/addition_rnn.pybabel
karpathy/neuraltalk · GitHub :自動根據圖像生成文本描述。
https://github.com/karpathy/neuraltalk網絡
ryankiros/neural-storyteller · GitHub: 看圖講故事
https://github.com/ryankiros/neural-storyteller架構
karpathy/neuraltalk2 · GitHub:看圖生成標註
https://github.com/karpathy/neuraltalk2
jcjohnson/neural-style · GitHub:將照片變成大師風格的繪畫
https://github.com/jcjohnson/neural-style
Newmu/dcgan_code · GitHub: 卷積生成式對抗網絡,生成圖像
https://github.com/Newmu/dcgan_code
nagadomi/waifu2x · GitHub:CNN來放大動漫圖片
https://github.com/nagadomi/waifu2x
去年我在Neuraltalk2 的基礎上作了個視頻字幕自動生成的實驗, 如今把代碼公佈在Github上:
GitHub - cgq5/Video-Caption-with-Neuraltalk2: Code release of captioning videos using Neuraltalk2.
https://github.com/cgq5/Video-Caption-with-Neuraltalk2