IdeaHack-SpamIdentifier

Web Portal

Clickbait Checker

Web application to check if the news headline is a clickbait or not and deciding wheather a message is spam.

Requirements

• Python modules (- pandas, - numpy, - scipy, - scikit-learn, - pickle, - flask, - pdb, - sys, - os)

Running the App

********************************************************* < Optional > *********************************************************

You may want to create a virtual environment as follows:

1. Create a virtual environment : virtualenv <YOUR_VIRTUALENV_NAME>
2. Activate virtual environment : source <YOUR_VIRTUALENV_NAME>/bin/activate - Linux <YOUR_VIRTUALENV_NAME>\Scripts\activate - Windows

********************************************************* </ Optional > ********************************************************

Installing Dependencies (Ignore if already done)

      pip install jupyter pandas numpy scipy sklearn
      (jupyter is optional)
      pip intsall pickle
      pip install Flask
      (pdb, sys and os modules installs by default while installing python, in case not you can use pip to install them)

To run the app navigate to the project folder in bash and run the following command :

python main.py

This will run the app at http://127.0.0.1:8083

The Application

The model used for prediction of clickbait used a corpus of clickbait and non-cickbait data aggregated from various sources.
The clickbait-detection.ipnyb shows the model training and its accuracy. The clickbaitmodelsklearn.pkl file stores the serialised object which is used for prediction.
The main.py and runner.py files host the flask app and score the headline in runtime.