Sentiment Analysis - API

Used the airline sentiment analysis dataset, which contained statements and their binary sentiments, i.e., positive and negative classes.

Tools Required

• Pandas - Python data manipulation libraries

• NLTK - Working with textual data

• Scikit-Learn - Vectorizors, Evaluation metrics, ML models

• Tensorflow - Deep Learning Networks

• BERT - Encoder Model

Roadmap

1. Main.ipynb This contains the model generated using the OOPs function.

2. Pipeline

• Installing libraries and dependency
• Importing the dataset
• Data Preprocessing - Basic preprocessing and cleaning the dataset
• Performing Vectorization and TfidfTransformer in the pipeline
• Dividing the dataset into train and test
• Applying Machine Learning models
  • Logistic Regression
  • XGB Classifier
  • Random Forest Classifier
  • SVC
• Applying Deep Learning models (Code not available now used for Research work)
• Generate pkl files and use them in API implementaion.
• Download the code file to your device and run the API.py to get a local server link for the API.

Report

Report

Contains the dataset description and comparisions between different ML/DL models.

Feedback

If you have any feedback, please reach out to us at [email protected]

🔗 Links