awesome-sentiment-analysis
Repository with all what is necessary for sentiment analysis and related areas
https://github.com/laugustyniak/awesome-sentiment-analysis
Last synced: 30 days ago
JSON representation
API
Lexicon-based Ensembles
Books
Lexicon-based Ensembles
- Sentiment Analysis: mining sentiments, opinions, and emotions - This book is suitable for students, researchers, and practitioners interested in natural language processing in general, and sentiment analysis, opinion mining, emotion analysis, debate analysis, and intention mining in specific. Lecturers can use the book in class.
Demos
Lexicon-based Ensembles
- Sentiment TreeBank - demo of Stanford's Treebank Sentiment Analysis
- NLTK Demo
- GATE Brexit Analyzer
- FormTitan
- Sentiment TreeBank - demo of Stanford's Treebank Sentiment Analysis
Explainable Sentiment Analysis
Methods & Tools
Practical Guides
- SHAP + LIME for Real-Time Predictions in Production
- DataCamp: Explainable AI with LIME, SHAP, and InterpretML
Survey Papers
Libraries
Aspect-based Sentiment Analysis
- Sentiment Analysis with Social Attention - Code for the TACL paper Overcoming Language Variation in Sentiment Analysis with Social Attention
- Twitter-sent-dnn - Deep Neural Network for Sentiment Analysis on Twitter.
- Aspect Based Sentiment Analysis - System that participated in Semeval 2014 task 4: Aspect Based Sentiment Analysis.
- Aspect Based Sentiment Analysis using End-to-End Memory Networks - TensorFlow implementation of Tang et al.'s EMNLP 2016 work.
- Generating Reviews and Discovering Sentiment - Code for Learning to Generate Reviews and Discovering Sentiment (Alec Radford, Rafal Jozefowicz, Ilya Sutskever).
- Neural Sentiment Classification - Neural Sentiment Classification aims to classify the sentiment in a document with neural models, which has been the state-of-the-art methods for sentiment classification. In this project, we provide our implementations of NSC, NSC+LA and NSC+UPA [Chen et al., 2016] in which user and product information is considered via attentions over different semantic levels.
- Helium MCP - time news with 37-dimension bias scoring, ML options pricing, and live market data. Interactive demo · REST API
Modern Transformer-based Libraries (2023-2026)
- Python, cardiffnlp/twitter-roberta-base-sentiment-latest - RoBERTa model fine-tuned for Twitter sentiment analysis, achieving state-of-the-art performance on social media text (updated 2024).
- Python, ModernFinBERT - Financial sentiment analysis model based on ModernBERT architecture (released July 2025), specialized for financial texts, earnings calls, and analyst reports, and reporting improved performance over earlier FinBERT variants on multiple financial sentiment benchmarks (see model card for details).
- Python, tabularisai/multilingual-sentiment-analysis - Multilingual sentiment analysis project targeting support for multiple languages; see the Hugging Face page for current availability and details.
- Python, Flair - Modern NLP framework with multilingual support and state-of-the-art sentiment analysis models, particularly strong for cross-lingual tasks.
- Python, Stanza - Stanford NLP library with multilingual support for 60+ languages, includes sentiment analysis capabilities.
Traditional Libraries
- Dragon Sentiment Classifier C# - Dragon Sentiment API is a C# implementation of the Naive Bayes Sentiment Classifier to analyze the sentiment of a text corpus.
- Software, RapidMiner - software capable of solving almost any text processing problem. processing text using computational linguistics.
- Python, Spacy - Industrial-Strength Natural Language Processing in Python, one of the best and the fastest libs for NLP. spaCy excels at large-scale information extraction tasks. It's written from the ground up in carefully memory-managed Cython. Independent research has confirmed that spaCy is the fastest in the world. If your application needs to process entire web dumps, spaCy is the library you want to be using.
- Python, TextBlob - TextBlob allows you to specify which algorithms you want to use under the hood of its simple API.
- Python, pattern - The pattern.en module contains a fast part-of-speech tagger for English (identifies nouns, adjectives, verbs, etc. in a sentence), sentiment analysis, tools for English verb conjugation and noun singularization & pluralization, and a WordNet interface.
- Software, GATE - GATE is open source software capable of solving almost any text processing problem.
- JAVA, OpenNLP - The Apache OpenNLP library is a machine learning based toolkit for the processing of natural language text.
- sentiment: Tools for Sentiment Analysis in R - sentiment is an R package with tools for sentiment analysis including bayesian classifiers for positivity/negativity and emotion classification.
- ASUM Java - Aspect and Sentiment Unification Model for Online Review Analysis.
- Software, RapidMiner - software capable of solving almost any text processing problem. processing text using computational linguistics.
- Software, KNIME - KNIME® Analytics Platform is the leading open solution for data-driven innovation, helping you discover the potential hidden in your data, mine for fresh insights, or predict new futures. Our enterprise-grade, open source platform is fast to deploy, easy to scale and intuitive to learn. With more than 1000 modules, hundreds of ready-to-run examples, a comprehensive range of integrated tools, and the widest choice of advanced algorithms available, KNIME Analytics Platform is the perfect toolbox for any data scientist. Our steady course on unrestricted open source is your passport to a global community of data scientists, their expertise, and their active contributions.
- Python, Textlytics - set of sentiment analysis examples based on Amazon Data, SemEval, IMDB etc.
- Java, Polish Sentiment Model - Sentiment analysis for polish language using SVM and BoW - within Docker.
- Java, LingPipe - LingPipe is tool kit for processing text using computational linguistics.
- C++, MITIE - MIT Information Extraction.
- Dragon Sentiment Classifier C# - Dragon Sentiment API is a C# implementation of the Naive Bayes Sentiment Classifier to analyze the sentiment of a text corpus.
- AFINN-based sentiment analysis for Node.js - Sentiment is a Node.js module that uses the AFINN-165 wordlist and Emoji Sentiment Ranking to perform sentiment analysis on arbitrary blocks of input text.
- SentiMental - Putting the Mental in Sentimental in js - Sentiment analysis tool for node.js based on the AFINN-111 wordlist. Version 1.0 introduces performance improvements making it both the first, and now fastest, AFINN backed Sentiment Analysis tool for node.
LLM Evaluation & Benchmarks for Sentiment Analysis
Benchmark Frameworks
- [paper
- [proceedings
- [call for participation
- [task paper - 1.452) [github] [DimStance datasets paper] [call for participation]
Explainable Sentiment Analysis Dataset
- Explainable Sentiment Analysis Dataset - truth sentiment labels, model predictions (GPT-4o, GPT-4o-mini, DeepSeek-R1), and fine-grained classifications for explainability evaluation.
Model Performance Leaderboard (2025-2026)
LLM Reliability & Safety in Sentiment Analysis
Bias & Fairness
- Bias in Large Language Models: Origin, Evaluation, and Mitigation
- Towards Trustworthy LLMs: Debiasing and Dehallucinating
- Towards Trustworthy LLMs: Debiasing and Dehallucinating
- Towards Trustworthy LLMs: Debiasing and Dehallucinating
- Towards Trustworthy LLMs: Debiasing and Dehallucinating
- Towards Trustworthy LLMs: Debiasing and Dehallucinating
Domain Instability
Hallucination
International Workshops
- [site
- [proceedings
- [2026 tasks - frameworks))
Uncertainty & Variability
LLM Techniques for Sentiment Analysis
In-Context Learning & Few-Shot Methods
Instruction Tuning & Alignment
Parameter-Efficient Fine-Tuning (PEFT)
- [paper
- PEFT Library
- [paper
- [paper
- QLoRA Repository
- [paper
- [paper
- QLoRA Official Guide
- 2025 Guide to Fine-Tuning: LoRA, QLoRA & Transfer Learning
- Keras: LoRA and QLoRA fine-tuning of Gemma
- [paper
- [paper
- [paper
- [paper
Prompt Engineering
- [source
- [paper
- [paper
- [paper
- [paper
- [paper
- [paper
- Prompt Engineering Guide — Sentiment Classification
- OpenAI Prompt Engineering Guide
- Comprehensive Taxonomy of Prompt Engineering Techniques
- [paper
- Comprehensive Taxonomy of Prompt Engineering Techniques
- [paper
- Comprehensive Taxonomy of Prompt Engineering Techniques
- [paper
- Comprehensive Taxonomy of Prompt Engineering Techniques
- [paper
- Comprehensive Taxonomy of Prompt Engineering Techniques
Reasoning Models for Sentiment Analysis
Retrieval-Augmented Generation (RAG)
- [paper
- LangChain
- LlamaIndex - heavy sentiment pipelines.
- LangGraph - based workflow orchestration for multi-step and agentic sentiment reasoning (2025).
- [arXiv 2504.14891
Multilingual and Cross-lingual Sentiment Analysis
Recent Approaches and Techniques
State-of-the-Art Models (2024-2025)
Multimodal Sentiment Analysis
Key Research Findings (2024-2025)
Recent Models and Frameworks (2024-2025)
Recent Models and Frameworks (2024-2026)
Papers
Aspect-Based Sentiment Analysis (2024-2025)
- A systematic review of aspect-based sentiment analysis: domains, methods, and trends - Comprehensive systematic review of ABSA methods and trends (2024)
- Large-Scale Aspect-Based Sentiment Analysis with Reasoning-Infused LLMs - Incorporating reasoning techniques into LLMs for ABSA (2025)
- Aspect-based Sentiment Analysis via Synthetic Image Generation - Novel approach generating sentimental images for ABSA (EMNLP 2025)
- Unifying aspect-based sentiment analysis BERT and multi-layered graph convolutional networks for comprehensive sentiment dissection - Multi-layered Enhanced Graph Convolutional Networks (MLEGCN) for ABSA (2024)
- Triple dimensional psychology knowledge encouraging graph attention networks to exploit aspect-based sentiment analysis - Psychology-informed graph attention networks (VADGAT) for ABSA (2025)
- Local interpretation of deep learning models for Aspect-Based Sentiment Analysis - Addressing interpretability in deep learning ABSA models (2025)
- A systematic review of aspect-based sentiment analysis: domains, methods, and trends - Comprehensive systematic review of ABSA methods and trends (2024)
- Unifying aspect-based sentiment analysis BERT and multi-layered graph convolutional networks for comprehensive sentiment dissection - Multi-layered Enhanced Graph Convolutional Networks (MLEGCN) for ABSA (2024)
- Triple dimensional psychology knowledge encouraging graph attention networks to exploit aspect-based sentiment analysis - Psychology-informed graph attention networks (VADGAT) for ABSA (2025)
- A systematic review of aspect-based sentiment analysis: domains, methods, and trends - Comprehensive systematic review of ABSA methods and trends (2024)
- Unifying aspect-based sentiment analysis BERT and multi-layered graph convolutional networks for comprehensive sentiment dissection - Multi-layered Enhanced Graph Convolutional Networks (MLEGCN) for ABSA (2024)
- Triple dimensional psychology knowledge encouraging graph attention networks to exploit aspect-based sentiment analysis - Psychology-informed graph attention networks (VADGAT) for ABSA (2025)
Aspect-Based Sentiment Analysis (2024-2026)
- Arctic-ABSA: Large-Scale Aspect-Based Sentiment Analysis with Reasoning-Infused LLMs - Snowflake's collection of ABSA models trained on public plus synthetic data (~20× larger than SemEval14), injecting reasoning via Chain-of-Thought fine-tuning and a novel reasoning-pretraining technique for encoder-only models; the 395M ModernBERT-large encoder and 8B decoder reportedly beat GPT-4o and Claude 3.5 Sonnet by up to 10 accuracy points, setting new SOTA on SemEval14 (91.8% restaurants / 87.2% laptops) (arXiv, January 2026)
- Zero-Shot Cross-Domain Aspect-Based Sentiment Analysis via Domain-Contextualized Chain-of-Thought Reasoning - ZeroABSA: unified zero-shot framework eliminating the need for labeled target-domain data via hybrid LLM-based data augmentation (filtered by vocabulary richness, semantic coherence, and sentiment/domain consistency) plus domain-aware chain-of-thought prompting (Findings of EMNLP 2025)
- Zero-Shot to Full-Resource: Cross-lingual Transfer Strategies for Aspect-Based Sentiment Analysis - Evaluation across seven languages (English, German, French, Dutch, Russian, Spanish, Czech) and four subtasks (ACD, ACSA, TASD, ASQP) under zero-resource, data-only, and full-resource settings; fine-tuned LLMs score highest overall, and the best cross-lingual transfer strategy is architecture-specific. Contributes two new German ABSA datasets, including GERest — the first German ASQP dataset (LREC 2026)
- Multi-domain Multilingual Sentiment Analysis in Industry: Predicting Aspect-based Opinion Quadruples - White & Shimorina. Industry-deployed system for aspect-based opinion quadruple extraction (aspect category, polarity, target, opinion expression) across multiple domains and languages; a single fine-tuned multi-domain LLM matches specialized single-domain models while reducing operational complexity, with practical guidance on handling non-extractive (hallucinated) predictions (2025)
- A systematic review of aspect-based sentiment analysis: domains, methods, and trends - Comprehensive systematic review of ABSA methods and trends (2024)
- Unifying aspect-based sentiment analysis BERT and multi-layered graph convolutional networks for comprehensive sentiment dissection - Multi-layered Enhanced Graph Convolutional Networks (MLEGCN) for ABSA (2024)
- Triple dimensional psychology knowledge encouraging graph attention networks to exploit aspect-based sentiment analysis - Psychology-informed graph attention networks (VADGAT) for ABSA (2025)
- A systematic review of aspect-based sentiment analysis: domains, methods, and trends - Comprehensive systematic review of ABSA methods and trends (2024)
- Unifying aspect-based sentiment analysis BERT and multi-layered graph convolutional networks for comprehensive sentiment dissection - Multi-layered Enhanced Graph Convolutional Networks (MLEGCN) for ABSA (2024)
- Triple dimensional psychology knowledge encouraging graph attention networks to exploit aspect-based sentiment analysis - Psychology-informed graph attention networks (VADGAT) for ABSA (2025)
Domain-Specific Applications (2024-2025)
- Analyzing student mental health with RoBERTa-Large: a sentiment analysis and data analytics approach - Healthcare application for mental health monitoring (2025)
- Comparative analysis of transformer models for sentiment classification of UK CBDC discourse on X - Financial sentiment analysis on social media (2025)
- Comparative analysis of transformer models for sentiment classification of UK CBDC discourse on X - Financial sentiment analysis on social media (2025)
- Comparative analysis of transformer models for sentiment classification of UK CBDC discourse on X - Financial sentiment analysis on social media (2025)
Domain-Specific Applications (2024-2026)
- Fine-tuning of lightweight large language models for sentiment classification on heterogeneous financial textual data - Benchmarks FinBERT against DeepSeek-LLM 7B, Llama3 8B Instruct, and Qwen3 8B on five public financial datasets (FinancialPhraseBank, FiQA, Gold News, Twitter, Chinese Finance Sentiment); fine-tuned Qwen3 8B and Llama3 8B perform best in most scenarios even with only 5% of training data, positioning lightweight open LLMs as cost-effective FinBERT alternatives (2025)
- Comparative analysis of transformer models for sentiment classification of UK CBDC discourse on X - Financial sentiment analysis on social media (2025)
- Comparative analysis of transformer models for sentiment classification of UK CBDC discourse on X - Financial sentiment analysis on social media (2025)
Language Models
- Sentiment Analysis in the Era of Large Language Models: A Reality Check -
- XLNet: Generalized Autoregressive Pretraining for Language Understanding -
- How to Fine-Tune BERT for Text Classification? -
- XLNet: Generalized Autoregressive Pretraining for Language Understanding -
- BERT
- Sentiment Analysis in the Era of Large Language Models: A Reality Check -
- How to Fine-Tune BERT for Text Classification? -
- Reasoning Implicit Sentiment with Chain-of-Thought Prompting -
Lexicon-based Ensembles
- Simpler is better? Lexicon-based ensemble sentiment classification beats supervised methods - lexicon-based ensemble can beat supervised learning.
- Comprehensive Study on Lexicon-based Ensemble Classification Sentiment Analysis - Comparison of several lexicon, supervised learning and ensemble methods for sentiment analysis.
Multilingual and Cross-lingual Sentiment Analysis (2024-2025)
- Multilingual sentiment analysis in restaurant reviews using aspect focused learning - XLM-RSA achieving 91.9% accuracy on multilingual restaurant reviews (2025)
- The Model Arena for Cross-lingual Sentiment Analysis: A Comparative Study in the Era of Large Language Models - Comparative study of LLMs vs SMLMs in cross-lingual settings (2024)
- Multilingual sentiment analysis in restaurant reviews using aspect focused learning - XLM-RSA achieving 91.9% accuracy on multilingual restaurant reviews (2025)
- Multilingual sentiment analysis in restaurant reviews using aspect focused learning - XLM-RSA achieving 91.9% accuracy on multilingual restaurant reviews (2025)