Sentiment Analysis

  1. Machine Learning Approaches
    1. Supervised Learning Framework
      1. Problem Formulation
        1. Classification vs Regression
          1. Binary vs Multi-class Classification
            1. Ordinal Classification
            2. Data Requirements
              1. Labeled Dataset Creation
                1. Annotation Guidelines
                  1. Quality Control
                  2. Training Process
                    1. Data Splitting
                      1. Model Training
                        1. Hyperparameter Tuning
                          1. Model Selection
                        2. Classical Machine Learning Models
                          1. Naive Bayes Classifiers
                            1. Multinomial Naive Bayes
                              1. Bernoulli Naive Bayes
                                1. Gaussian Naive Bayes
                                  1. Feature Independence Assumption
                                  2. Logistic Regression
                                    1. Linear Models
                                      1. Regularization Techniques
                                        1. Feature Scaling
                                        2. Support Vector Machines
                                          1. Linear SVM
                                            1. Kernel Methods
                                              1. Soft Margin Classification
                                                1. Parameter Tuning
                                                2. Tree-Based Methods
                                                  1. Decision Trees
                                                    1. Random Forests
                                                      1. Gradient Boosting
                                                        1. Feature Importance
                                                        2. Ensemble Methods
                                                          1. Bagging
                                                            1. Boosting
                                                              1. Stacking
                                                                1. Voting Classifiers
                                                              2. Unsupervised Learning Methods
                                                                1. Clustering Approaches
                                                                  1. K-means Clustering
                                                                    1. Hierarchical Clustering
                                                                      1. DBSCAN
                                                                        1. Gaussian Mixture Models
                                                                        2. Topic Modeling
                                                                          1. Latent Semantic Analysis
                                                                            1. Latent Dirichlet Allocation
                                                                              1. Non-negative Matrix Factorization
                                                                                1. Topic-Based Sentiment
                                                                                2. Dimensionality Reduction
                                                                                  1. Principal Component Analysis
                                                                                    1. t-SNE
                                                                                      1. UMAP
                                                                                    2. Semi-Supervised Learning
                                                                                      1. Self-Training
                                                                                        1. Co-Training
                                                                                          1. Label Propagation
                                                                                            1. Pseudo-Labeling
                                                                                              1. Active Learning
                                                                                              2. Hybrid Approaches
                                                                                                1. Lexicon-ML Combinations
                                                                                                  1. Ensemble of Different Approaches
                                                                                                    1. Rule-Based Post-Processing

                                                                                                  Previous

                                                                                                  3. Lexicon-Based Sentiment Analysis

                                                                                                  Go to top

                                                                                                  Next

                                                                                                  5. Feature Engineering and Representation