BSski / 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.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool



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

πŸ“œ 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.

πŸ”¨ 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.

πŸ“ž Contact

πŸ‘· Author

πŸ”“ License

MIT

About

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


Languages

Language:Jupyter Notebook 100.0%