/classifiers-comparative-analysis

01.2021: Analysis of performance of SVC, linear regression and neural network classifiers in recognizing premise-conclusion pairs of sentences based on semantic similarity.

Primary LanguageJupyter Notebook


WARNING: VERY OLD CODE

This code was written in January of 2021. I have learned a lot since then and I am aware of the poor quality of the code.

Comparative analysis of the performance of SVC, linear regression and artificial neural network classifiers in recognizing premise-conclusion pairs and random pairs of sentences based on semantic similarity

Table of contents

📜 Project description

Analysis of performance of SVC, linear regression and neural network classifiers in recognizing premise-conclusion pairs of sentences based on semantic similarity. All classifiers achieved accuracy of 79%+/-0.5%. Performance of the classifiers was limited by uni-dimensionality of the data.

🔨 Technologies used

  • Python
  • Numpy
  • Pandas
  • Scikit-learn
  • Tensorflow
  • Keras
  • Matplotlib

⬆️ Room for improvement

This is an old project of mine and it would certainly benefit from cleaning the code and a general refactoring.

📞 Contact

👷 Author

🔓 License

MIT