/Text-Classification-Topsis

This project uses the topsis package to determine the best model for text classification (NLP)

Primary LanguageJupyter NotebookMIT LicenseMIT

Evaluation of Text Classification Pretrained Models using Topsis Package

License

Bargraph showing topsis score vs models


Evaluation parameters of different models

TOPSIS Score Calculation

Utilized the custom 'Topsis-Divyam-102103142' package to compute TOPSIS scores based on the model performance metrics. Link to Topsis-Divyam-10210342 Package

Key Features

  • Model Selection and Integration: Incorporates popular pretrained models, such as BERT and GPT, to capture intricate patterns and semantic nuances within textual data.

Evaluation Metrics

  • Accuracy
  • Precision (Macro and Micro)
  • Recall (Macro and Micro)
  • F1 Score (Macro and Micro)
  • Cohen's Kappa

Use Cases

  • Sentiment Analysis
  • Topic Classification
  • Document Categorization
  • Spam Detection
  • Customer Feedback Analysis

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

Divyam Malik

Github Link: github