Sentiment Analysis
情感分析是自然语言处理领域的一项任务,旨在对给定文本的情感倾向进行分类,通常将其归类为正面、负面或中性。该任务通过机器学习、基于词典的方法及混合方法实现,近年来深度学习技术如RoBERTa和T5被广泛应用于训练高性能情感分类器,评估指标包括F1分数、召回率和精确率。情感分析不仅用于社交媒体监控,还广泛应用于产品评论分析、市场趋势预测等领域,具有重要的应用价值。
1B Words
AJGT
AraBERTv1
Amazon Review Full
BERT large
Amazon Review Polarity
BERT large
ArSAS
ASTD
BanglaBook
Bangla-BERT (large)
ChnSentiCorp
ChnSentiCorp Dev
CR
AnglE-LLaMA-7B
DBRD
RobBERT
DynaSent
SVM
Financial PhraseBank
FiLM
FiQA
HARD
IITP Movie Reviews Sentiment
IITP Product Reviews Sentiment
CalBERT
IMDb
RoBERTa-large with LlamBERT
IMDb Movie Reviews
Space-XLNet
LABR (2-class, unbalanced)
Latvian Twitter Eater Sentiment Dataset
Naive Bayes
MPQA
MR
VLAWE
Multi-Domain Sentiment Dataset
UDALM: Unsupervised Domain Adaptation through Language Modeling
RuSentiment
RuBERT-RuSentiment
SAIL 2017
SemEval
SemEval 2014 Task 4 Subtask 1+2
SemEval 2017 Task 4-A
LSTMs+CNNs ensemble with multiple conv. ops
Sentiment Merged
GPT-4o Fine-Tuned (Minimal)
SLUE
Sogou News
fastText, h=10, bigram
SST-2 Binary classification
T5-11B
SST-3
SST-5 Fine-grained classification
Heinsen Routing + RoBERTa Large
TweetEval
BERTweet
Twitter
AEN-BERT
Urdu Online Reviews
RCNN
User and product information
MA-BERT
Yelp Binary classification
XLNet
Yelp Fine-grained classification
XLNet
lstm+bert