prateek025 / nlp-using-pycaret

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Natural Language Processing Text Classification : PyCaret Approach Vs The Traditional Approach

  • This repository contains a tutorial, Spam-Ham.ipynb, that covers solving a text classification problem by building machine learning models in python.
  • SMS Spam Collection Data Set has been taken from UCI Machine Learning Repository for model development. Different approaches and different binary classification models were developed on this dataset to identify if a given text message a Spam message or a Ham message
  • One approach was to use python package PyCaret, a low-code ML library, to build and tune machine learning models with minimum lines of code. The other approach was the traditional way of using scikit-learn and other phtyhon libraries to solve the same problem. A quick comparison between the two approaches has been recorded in the following table.
  • More step-wise details on this code and approach can be found on my Medium article.

About


Languages

Language:Jupyter Notebook 100.0%