title	Smsfilter
emoji	🚀
colorFrom	red
colorTo	red
sdk	docker
app_port	8501
tags	streamlit
pinned	false
short_description	Streamlit template space

SMS Spam Classification

A python jupyter notebook which will train machine learning models to detect Spam messages.The project is hosted at streamlit app on the web.

Features

• Data cleaning.
• Data preprocessing.
• Extensive data analysis.
• Text vectorization using Bag of words , Tf-idf , Word2Vec encoding.
• 8 classification model training using pipeline.
• Above 95% accurate and precise.
• Hosting using Streamlit.

Prerequisite

You need python installed in your system to run this repository.

Installation

Install the dependencies from the requirements.txt.

pip install -r requirements.txt

To host the project locally , run app.py with the help of streamlit.

streamlit run app.py

Drawbacks

The project has some drawbacks.

• Does not have 100% accuracy and percision.
• Model building requires lot of memory (1.5GB).
• Longer training period.
• Might throw exceptions.

Support

• Email : pratham20442@gamil.com
• github : https://github.com/pratham-ak2004
• linkedin : https://www.linkedin.com/in/pratham-a-kadekar-8397a7249/