https://github.com/gauravkaklotar/emotions_analysis

Emotions Analysis
https://github.com/gauravkaklotar/emotions_analysis

data-analyst flask-api ml nlkt python3 reactjs tailwindcss

Last synced: about 2 months ago
JSON representation

Emotions Analysis

• Host: GitHub
• URL: https://github.com/gauravkaklotar/emotions_analysis
• Owner: GauravKaklotar
• License: mit
• Created: 2024-08-12T06:47:17.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2024-08-29T11:31:04.000Z (6 months ago)
• Last Synced: 2024-11-08T17:49:11.543Z (3 months ago)
• Topics: data-analyst, flask-api, ml, nlkt, python3, reactjs, tailwindcss
• Language: Jupyter Notebook
• Homepage:
• Size: 18.2 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Emotion Detection from Text

This project focuses on classifying emotions in Twitter text data into six categories: sadness, joy, love, anger, fear, and surprise. The dataset consists of Twitter messages, each labeled with one of these emotions.

## Project Overview

1. **Data Preprocessing**:

   - Loaded the Twitter text dataset and performed text cleaning.

   - Removed URLs, mentions, hashtags, and special characters from the text.

   - Tokenized and normalized the text for consistent analysis.

2. **Exploratory Data Analysis (EDA)**:

   - Analyzed the distribution of emotions in the dataset.

   - Examined the length of text segments to understand the data structure.

   - Generated visualizations, including bar plots and histograms, to gain insights into the dataset.

3. **Model Selection and Training**:

   - Chose Logistic Regression for initial model training.

   - Vectorized the text data using TF-IDF to convert it into numerical features.

   - Split the dataset into training and testing sets with an 80/20 ratio.

   - Trained the model on the training data and evaluated its performance.

4. **Model Evaluation**:

   - Assessed the model using accuracy, precision, recall, and F1-score metrics.

   - Generated a classification report to evaluate the performance across different emotion categories.

   - Created a confusion matrix to visualize the model’s prediction accuracy.

5. **Model Improvement**:

   - Performed hyperparameter tuning using Grid Search to optimize the Logistic Regression model.

   - Validated the model using cross-validation to ensure it generalizes well to unseen data.