sooftware / nlp-tasks

Natural Language Processing Tasks and Examples.

Natural Language Processing Tasks and Examples

With the advancement of A.I. technology in recent years, natural language processing technology has been able to solve so many problems. While working as an NLP engineer, I encountered various tasks, and I thought it would be nice to gather and organize the natural language processing tasks I have dealt with in one place. Borrowing Kyubyong's project format, I organized natural language processing tasks with references and example code.

Automated Essay Scoring

• WIKI Automated Essay Scoring
• DATA The Hewlett Foundation: Automated Essay Scoring
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's AES

Automatic Speech Recognition

• WIKI Speech Recognition
• DATA LibriSpeech
• DATA AISHELL-1
• DATA KsponSpeech
• MODEL Deep Speech2
• MODEL Listen, Attend and Spell
• MODEL Wav2vec 2.0
• OFF-THE-SHELF Pororo's ASR
• CODE Example with KsponSpeech

Dialogue Generation

• WIKI Dialogue System
• DATA Persona Chat
• DATA Korean SNS Corpus
• MODEL Dialogue GPT
• CODE Example with Korean SNS Corpus

Dialogue Retrieval

• WIKI Dialogue System
• DATA Persona Chat
  DATA Ubuntu Dialogue Corpus
• DATA Korean SNS Corpus
• MODEL Poly Encoder
• CODE Example with Ubuntu Dialogue Corpus

Fill in the Blank

• WIKI Cloze Test
• INFO Masked-Language-Modeling with BERT
• MODEL BERT
• MODEL RoBERTa
• OFF-THE-SHELF Pororo's Fill in the Blank
• CODE Example with WikiCorpus

Grammatical Error Correction

• WIKI Autocorrection
• DATA NUS Non-commercial research/trial corpus license
• DATA Cornell Movie--Dialogs Corpus
• OFF-THE-SHELF Pororo's GEC

Grapheme To Phoneme

• WIKI Grapheme
• WIKI Phoneme
• REPRESENTATIVE-DATA Multilingual Pronunciation Data
• OFF-THE-SHELF-MODEL Pororo's G2P

Knowledge Powered Conversation

• PAPER Wizard of Wikipedia: Knowledge-Powered Conversational agents
• DATA Wizard of Wikipedia
• CODE Example with Wizard of Wikipedia

Language Modeling

• WIKI Language Model
• INFO A beginner’s guide to language models
• MODEL GPT3
• MODEL GPT2
• MODEL Ken-LM
• MODEL RNN-LM
• CODE Example with OpenWebText

Machine Reading Comprehension

• WIKI Reading Comprehension
• INFO Machine Reading Comprehension with BERT
• DATA SQuAD
• DATA KorQuad
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's MRC
• CODE Example with SQuAD & KorQuad

Machine Translation

• WIKI Translation
• DATA WMT 2014 English-to-French
• DATA Korean-English translation corpus
• MODEL Transformer
• OFF-THE-SHELF Pororo's Translation
• CODE Example with Korean-English translation corpus

Math Word Problem Solving

• PAPER-WITH-CODE Math Word Problem Solving
• DATA DeepMind Mathmatics Dataset
• DATA KMWP (Korean Math Word Problems)
• CODE Example with KMWP

Natural Language Inference

• WIKI Textual Entailment
• DATA GLUE-MNLI
• DATA KorNLI
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's NLI
• CODE Example with GLUE-MNLI

Named Entity Recognition

• WIKI Named Entity Recognition
• DATA CoNLL-2002 NER corpus
• DATA CoNLL-2003 NER corpus
• DATA Naver NER
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's NER
• CODE Example with Naver NER

Paraphrase Generation

• WIKI Paraphrase
• OFF-THE-SHELF Pororo's Paraphrase Generation

Phoneme To Grapheme

• OFF-THE-SHELF Pororo's P2G

Sentiment Analysis

• WIKI Sentiment Analysis
• DATA GLUE-SST
• DATA NSMC
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's Sentiment Analysis
• CODE Example with NSMC

Semantic Textual Similarity

• WIKI Semantic Similarity
• DATA GLUE-STS
• DATA KorSTS
• MODEL BERT
• MODEL RoBERTa
• MODEL Electra
• OFF-THE-SHELF Pororo's STS
• CODE Example with SQuAD

Speech Synthesis

• WIKI Speech Synthesis
• DATA LJ Speech
• DATA CSS10
• DATA KSS
• MODEL Tacotron2
• MODEL FastSpeech2
• MODEL WaveNet
• MODEL Hifi-GAN
• OFF-THE-SHELF Pororo's TTS
• CODE Example with LJ-Speech
• CODE Example with KSS

Summarization

• WIKI Automatic Summarization
• DATA XSum
• DATA Korean Summarization Corpus
• MODEL BART
• OFF-THE-SHELF Pororo's Summarization
• CODE Example with XSum

Author

• Soohwan Kim @sooftware
• Contacts: [e-mail]