SOTAVerified

Sentiment Analysis

Sentiment Analysis is the task of classifying the polarity of a given text. For instance, a text-based tweet can be categorized into either "positive", "negative", or "neutral". Given the text and accompanying labels, a model can be trained to predict the correct sentiment.

Sentiment Analysis techniques can be categorized into machine learning approaches, lexicon-based approaches, and even hybrid methods. Some subcategories of research in sentiment analysis include: multimodal sentiment analysis, aspect-based sentiment analysis, fine-grained opinion analysis, language specific sentiment analysis.

More recently, deep learning techniques, such as RoBERTa and T5, are used to train high-performing sentiment classifiers that are evaluated using metrics like F1, recall, and precision. To evaluate sentiment analysis systems, benchmark datasets like SST, GLUE, and IMDB movie reviews are used.

Further readings:

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 4101–4150 of 5630 papers

TitleStatusHype
Data Sets: Word Embeddings Learned from Tweets and General Data—0
Leveraging Sparse and Dense Feature Combinations for Sentiment Classification—0
Radical-level Ideograph Encoder for RNN-based Sentiment Analysis of Chinese and Japanese—0
Mining fine-grained opinions on closed captions of YouTube videos with an attention-RNNCode0
LIA at SemEval-2017 Task 4: An Ensemble of Neural Networks for Sentiment Classification—0
Learning Word Representations with Regularization from Prior Knowledge—0
LSIS at SemEval-2017 Task 4: Using Adapted Sentiment Similarity Seed Words For English and Arabic Tweet Polarity Classification—0
Learning Stock Market Sentiment Lexicon and Sentiment-Oriented Word Vector from StockTwits—0
Learning Contextual Embeddings for Structural Semantic Similarity using Categorical Information—0
Learned in Translation: Contextualized Word VectorsCode0
MI\&T Lab at SemEval-2017 task 4: An Integrated Training Method of Word Vector for Sentiment Classification—0
INGEOTEC at SemEval 2017 Task 4: A B4MSA Ensemble based on Genetic Programming for Twitter Sentiment Analysis—0
INF-UFRGS at SemEval-2017 Task 5: A Supervised Identification of Sentiment Score in Tweets and Headlines—0
IITPB at SemEval-2017 Task 5: Sentiment Prediction in Financial Text—0
IITP at SemEval-2017 Task 5: An Ensemble of Deep Learning and Feature Based Models for Financial Sentiment Analysis—0
IBA-Sys at SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogs and News—0
NLG301 at SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogs and News—0
NNEMBs at SemEval-2017 Task 4: Neural Twitter Sentiment Classification: a Simple Ensemble Method with Different Embeddings—0
NRU-HSE at SemEval-2017 Task 4: Tweet Quantification Using Deep Learning Architecture—0
OMAM at SemEval-2017 Task 4: English Sentiment Analysis with Conditional Random Fields—0
OMAM at SemEval-2017 Task 4: Evaluation of English State-of-the-Art Sentiment Analysis Models for Arabic and a New Topic-based Model—0
HLP@UPenn at SemEval-2017 Task 4A: A simple, self-optimizing text classification system combining dense and sparse vectors—0
On the ``Calligraphy'' of Books—0
HHU at SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Data using Machine Learning Methods—0
HCS at SemEval-2017 Task 5: Polarity detection in business news using convolutional neural networks—0
Gradual Learning of Matrix-Space Models of Language for Sentiment Analysis—0
funSentiment at SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogs Using Word Vectors Built from StockTwits and Twitter—0
funSentiment at SemEval-2017 Task 4: Topic-Based Message Sentiment Classification by Exploiting Word Embeddings, Text Features and Target Contexts—0
Personality Driven Differences in Paraphrase Preference—0
Feature Selection as Causal Inference: Experiments with Text Classification—0
Exploring Joint Neural Model for Sentence Level Discourse Parsing and Sentiment Analysis—0
ELiRF-UPV at SemEval-2017 Task 4: Sentiment Analysis using Deep Learning—0
ej-sa-2017 at SemEval-2017 Task 4: Experiments for Target oriented Sentiment Analysis in Twitter—0
EICA at SemEval-2017 Task 4: A Simple Convolutional Neural Network for Topic-based Sentiment Classification—0
RiTUAL-UH at SemEval-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks—0
ECNU at SemEval-2017 Task 8: Rumour Evaluation Using Effective Features and Supervised Ensemble Models—0
ECNU at SemEval-2017 Task 5: An Ensemble of Regression Algorithms with Effective Features for Fine-Grained Sentiment Analysis in Financial Domain—0
ECNU at SemEval-2017 Task 4: Evaluating Effective Features on Machine Learning Methods for Twitter Message Polarity Classification—0
DUTH at SemEval-2017 Task 5: Sentiment Predictability in Financial Microblogging and News Articles—0
DUTH at SemEval-2017 Task 4: A Voting Classification Approach for Twitter Sentiment Analysis—0
Does the Geometry of Word Embeddings Help Document Classification? A Case Study on Persistent Homology-Based Representations—0
SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogs and News—0
SentiHeros at SemEval-2017 Task 5: An application of Sentiment Analysis on Financial Tweets—0
SentiME++ at SemEval-2017 Task 4: Stacking State-of-the-Art Classifiers to Enhance Sentiment Classification—0
deepSA at SemEval-2017 Task 4: Interpolated Deep Neural Networks for Sentiment Analysis in Twitter—0
DataStories at SemEval-2017 Task 4: Deep LSTM with Attention for Message-level and Topic-based Sentiment AnalysisCode0
CrystalNest at SemEval-2017 Task 4: Using Sarcasm Detection for Enhancing Sentiment Classification and Quantification—0
SINAI at SemEval-2017 Task 4: User based classification—0
COMMIT at SemEval-2017 Task 5: Ontology-based Method for Sentiment Analysis of Financial Headlines—0
SiTAKA at SemEval-2017 Task 4: Sentiment Analysis in Twitter Based on a Rich Set of Features—0
Show:102550
← PrevPage 83 of 113Next →
All datasetsSST-2 Binary classificationIMDbSST-5 Fine-grained classificationYelp Binary classificationMRYelp Fine-grained classificationBanglaBookDynaSentSST-3Sentiment MergedUser and product informationAmazon Review Full

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Word+ES (Scratch)Attack Success Rate100—Unverified
2MT-DNN-SMARTAccuracy97.5—Unverified
3T5-11BAccuracy97.5—Unverified
4MUPPET Roberta LargeAccuracy97.4—Unverified
5T5-3BAccuracy97.4—Unverified
6ALBERTAccuracy97.1—Unverified
7StructBERTRoBERTa ensembleAccuracy97.1—Unverified
8XLNet (single model)Accuracy97—Unverified
9SMARTRoBERTaDev Accuracy96.9—Unverified
10ELECTRAAccuracy96.9—Unverified
#ModelMetricClaimedVerifiedStatus
1RoBERTa-large with LlamBERTAccuracy96.68—Unverified
2RoBERTa-largeAccuracy96.54—Unverified
3XLNetAccuracy96.21—Unverified
4Heinsen Routing + RoBERTa LargeAccuracy96.2—Unverified
5RoBERTa-large 355M + Entailment as Few-shot LearnerAccuracy96.1—Unverified
6GraphStarAccuracy96—Unverified
7DV-ngrams-cosine with NB sub-sampling + RoBERTa.baseAccuracy95.94—Unverified
8DV-ngrams-cosine + RoBERTa.baseAccuracy95.92—Unverified
9Roberta_Large ST + Cosine Similarity LossAccuracy95.9—Unverified
10BERT large finetune UDAAccuracy95.8—Unverified
#ModelMetricClaimedVerifiedStatus
1Llama-3.3-70B + CAPOAccuracy62.27—Unverified
2Mistral-Small-24B + CAPOAccuracy 60.2—Unverified
3Heinsen Routing + RoBERTa LargeAccuracy59.8—Unverified
4RoBERTa-large+Self-ExplainingAccuracy59.1—Unverified
5Qwen2.5-32B + CAPOAccuracy 59.07—Unverified
6Heinsen Routing + GPT-2Accuracy58.5—Unverified
7BCN+Suffix BiLSTM-Tied+CoVeAccuracy56.2—Unverified
8BERT LargeAccuracy55.5—Unverified
9LM-CPPF RoBERTa-baseAccuracy54.9—Unverified
10BCN+ELMoAccuracy54.7—Unverified
#ModelMetricClaimedVerifiedStatus
1Char-level CNNError4.88—Unverified
2SVDCNNError4.74—Unverified
3LEAMError4.69—Unverified
4fastText, h=10, bigramError4.3—Unverified
5SWEM-hierError4.19—Unverified
6SRNNError3.96—Unverified
7M-ACNNError3.89—Unverified
8DNC+CUWError3.6—Unverified
9CCCapsNetError3.52—Unverified
10Block-sparse LSTMError3.27—Unverified
#ModelMetricClaimedVerifiedStatus
1Millions of EmojiTraining Time1,500—Unverified
2VLAWEAccuracy93.3—Unverified
3RoBERTa-large 355M + Entailment as Few-shot LearnerAccuracy92.5—Unverified
4AnglE-LLaMA-7BAccuracy91.09—Unverified
5byte mLSTM7Accuracy86.8—Unverified
6MEANAccuracy84.5—Unverified
7RNN-CapsuleAccuracy83.8—Unverified
8Capsule-BAccuracy82.3—Unverified
9SuBiLSTM-TiedAccuracy81.6—Unverified
10USE_T+CNNAccuracy81.59—Unverified