Thumbs up? Sentiment Classification using Machine Learning Techniques

Abstract

We consider the problem of classifying documents not by topic, but by overall sentiment, e.g., determining whether a review is positive or negative. Using movie reviews as data, we find that standard machine learning techniques definitively outperform human-produced baselines. However, the three machine learning methods we employed (Naive Bayes, maximum entropy classification, and support vector machines) do not perform as well on sentiment classification as on traditional topic-based categorization. We conclude by examining factors that make the sentiment classification problem more challenging.

To appear in EMNLP-2002

References in corpus (1)

• Automatic Detection of Text Genre

Cited by in corpus (51)

• From Frequency to Meaning: Vector Space Models of Semantics
• Wikipedia-based Semantic Interpretation for Natural Language Processing
• Empath: Understanding Topic Signals in Large-Scale Text
• A Survey on Aspect-Based Sentiment Classification
• Twitter Sentiment Analysis via Bi-sense Emoji Embedding and Attention-based LSTM
• Indonesian Social Media Sentiment Analysis With Sarcasm Detection
• Sentiment Analysis For Modern Standard Arabic And Colloquial
• Improving Twitter Sentiment Classification via Multi-Level Sentiment-Enriched Word Embeddings
• TensiStrength: Stress and relaxation magnitude detection for social media texts
• Tripartite Graph Clustering for Dynamic Sentiment Analysis on Social Media
• And That's A Fact: Distinguishing Factual and Emotional Argumentation in Online Dialogue
• Deep Learning-based Sentiment Classification: A Comparative Survey
• Sentiment Analysis : A Literature Survey
• Automatic Aggregation by Joint Modeling of Aspects and Values
• A Simple Regularization-based Algorithm for Learning Cross-Domain Word Embeddings
• Enhanced Twitter Sentiment Classification Using Contextual Information
• Sentiment Analysis on Bangla and Romanized Bangla Text (BRBT) using Deep Recurrent models
• The Royal Birth of 2013: Analysing and Visualising Public Sentiment in the UK Using Twitter
• Sentiment Classification of Customer Reviews about Automobiles in Roman Urdu
• An Architecture for Software Engineering Gamification
• Training-free Lexical Backdoor Attacks on Language Models
• Low-Resource Text Classification using Domain-Adversarial Learning
• Operational Calibration: Debugging Confidence Errors for DNNs in the Field
• Which Clustering Do You Want? Inducing Your Ideal Clustering with Minimal Feedback
• Structural block driven - enhanced convolutional neural representation for relation extraction
• Language Independent Sequence Labelling for Opinion Target Extraction
• Review Mining for Feature Based Opinion Summarization and Visualization
• KATE: K-Competitive Autoencoder for Text
• Efficient Path Prediction for Semi-Supervised and Weakly Supervised Hierarchical Text Classification
• Boost Phrase-level Polarity Labelling with Review-level Sentiment Classification
• Combination of Domain Knowledge and Deep Learning for Sentiment Analysis
• The role of entanglement for enhancing the efficiency of quantum kernels towards classification
• Persian Sentiment Analyzer: A Framework based on a Novel Feature Selection Method
• A Context-based Disambiguation Model for Sentiment Concepts Using a Bag-of-concepts Approach
• An Empirical Analysis of the Role of Amplifiers, Downtoners, and Negations in Emotion Classification in Microblogs
• Extracting Sentiment Attitudes From Analytical Texts
• Quality of Word Embeddings on Sentiment Analysis Tasks
• Idioms-Proverbs Lexicon for Modern Standard Arabic and Colloquial Sentiment Analysis
• Mining Software Quality from Software Reviews: Research Trends and Open Issues
• Combination of Domain Knowledge and Deep Learning for Sentiment Analysis of Short and Informal Messages on Social Media
• Ideogram Based Chinese Sentiment Word Orientation Computation
• Sentiment Analysis with Contextual Embeddings and Self-Attention
• Do sequence-to-sequence VAEs learn global features of sentences?
• Performance Investigation of Feature Selection Methods
• Weakly Supervised Reasoning by Neuro-Symbolic Approaches
• Neural Generators of Sparse Local Linear Models for Achieving both Accuracy and Interpretability
• Emotion Analysis Platform on Chinese Microblog
• Detecting Low Rating Android Apps Before They Have Reached the Market
• Computing Lexical Contrast
• A Convolutional Neural Network for Aspect Sentiment Classification
• Deep learning based mood tagging for Chinese song lyrics