HyperAI超神经

文本分类

Text Classification是自然语言处理中的核心任务,旨在将文本数据归类到预定义的类别中。该任务通过分析文本内容,识别其主题、情感或意图等特征,实现高效的信息组织与检索。近年来,深度学习模型如XLNet和RoBERTa显著提升了文本分类的性能,推动了技术的发展。Benchmark数据集如GLUE和AGNews被广泛用于评估模型的效果。

MTEB
ST5-XXL
AG News
DBpedia
XLNet
R8
RoBERTaGCN
TREC-6
Automatic Label Error Correction
20NEWS
RoBERTaGCN
UK Key Stage Readability
MR
Ohsumed
SGCN
Yahoo! Answers
BERT-ITPT-FiT
NewsDiscourse
R52
GraphStar
Yelp-5
HAHNN (CNN)
DODF Data
ULMFiT (pre-trained vocab, no gradual unfreezing)
Lot-insts
Character-BERT+RS
MVICTOR (type)
OneStopEnglish (Readability Assessment)
RoBERTa-RF-T1 hybrid
SVICTOR (type)
WeeBit (Readability Assessment)
BART-RF-T1 hybrid
Yelp-2
Amazon-2
arXiv-10
Protoformer
HateXplain
RCV1
NLP-Cap
ThreatGram 101 - Extreme Telegram Data
GPT-2
Amazon-5
BLURB
BioLinkBERT (large)
IMDb Movie Reviews
Logistic Regression
Overruling
Custom Legal-BERT
Sogou News
BERT-ITPT-FiT
Terms of Service
Twitter
An Amharic News Text classification Dataset
Naive Bayes using Tf-idf features
GLUE SST2
MuLD (Character Type)
Searchsnippets
Social media attributions of YouTube comments
SST-2
This is not a Dataset
TREC-50
20 Newsgroups
RoBERTaGCN
Adverse Drug Events (ADE) Corpus
AffCon 2020 Emotion Detection
Arxiv HEP-TH citation graph
BigBird
BANKING77
Facebook Media
FMC-MWO2KG
Flair
GLUE MRPC
GLUE RTE
Hyperpartisan News Detection
BigBird
Hyperpartisan
NICE-2
NICE-45
Patents
BigBird
RusAge: Corpus for Age-Based Text Classification
LSVC + linguistic features + publishing attributes
SILICONE Benchmark
STOPS-2
ERNIE 2.0
STOPS-41
TRAC2-Benghali. Task 2.
BERT
TRAC2-English. Task2.
TREC-10
BERT
Twitter Sentiment Analysis
Logistic Regression
Twitter-US
WNUT-2020 Task 2
NutCracker
ade_corpus_v2Ade_corpus_v2_classification
amazon_reviews_multi
book-text-classifier
catalonia_independence
clinc_oos
emotion
financial_phrasebank
GLUE
GLUE COLA
GLUE QQP
GLUE STSB
hate_speech18
IMDb
KLUE
MNIST
New_York_Times_Topics
NSFW-Safe-Dataset
SemEval 2014 Task 4 (Restaurants)
SST2
tecla
Unknown