https://github.com/hayatiyrtgl/flm_reviews_analysis

Sentiment analysis using ML classifiers for text data.
https://github.com/hayatiyrtgl/flm_reviews_analysis

countvectorizer machine-learning natural-language-processing natural-language-understanding nltk python python3 random-forest random-forest-classifier scikitlearn-machine-learning sentiment-analysis sentiment-classification sklearn sklearn-classify

Last synced: 4 months ago
JSON representation

Sentiment analysis using ML classifiers for text data.

• Host: GitHub
• URL: https://github.com/hayatiyrtgl/flm_reviews_analysis
• Owner: HayatiYrtgl
• License: mit
• Created: 2024-04-24T19:21:09.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-04-24T19:22:07.000Z (over 2 years ago)
• Last Synced: 2025-04-08T18:39:24.734Z (over 1 year ago)
• Topics: countvectorizer, machine-learning, natural-language-processing, natural-language-understanding, nltk, python, python3, random-forest, random-forest-classifier, scikitlearn-machine-learning, sentiment-analysis, sentiment-classification, sklearn, sklearn-classify
• Language: Jupyter Notebook
• Homepage:
• Size: 520 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
This script is a sentiment analysis project using machine learning classifiers such as Random Forests, Logistic Regression, Naive Bayes, K-Nearest Neighbors (KNN), and Support Vector Machines (SVM). Let's break down the key parts of the code:

1. **Data Loading and Preprocessing**:

   - The script loads training and test data from CSV files.

   - It checks for missing values in both datasets and converts text to lowercase.

   - Regular expressions are used to remove non-alphabetic characters from the text.

   - Text is split into words and stemmed using the Porter stemming algorithm.

   - Stop words are removed from the text.

2. **Feature Extraction**:

   - CountVectorizer is used to convert text data into numerical features.

3. **Model Training**:

   - Random Forests, Logistic Regression, Naive Bayes, KNN, and SVM classifiers are trained on the training data.

4. **Model Evaluation**:

   - Accuracy and F1 score are computed for each classifier using the test data.

5. **Prediction**:

   - There's a function `preprocess` defined to preprocess custom input.

   - Custom input is preprocessed and transformed using the trained CountVectorizer.

   - Predictions are made using each classifier for the custom input.

6. **Output**:

   - For the custom input, the script outputs the predicted probabilities (for Logistic Regression) and the predicted classes for each classifier.

Overall, this script provides a comprehensive sentiment analysis pipeline, from data preprocessing to model evaluation and prediction. It demonstrates the use of various machine learning classifiers for sentiment analysis tasks.