Skip to content

This app predicts the nature of the tweet into 6 Categories. i.e. Age, Ethnicity, Gender, Religion, Other Cyberbullying, Not Cyberbullying.

Notifications You must be signed in to change notification settings

apurvayadav/cyberbullying-tweet-recognition-app

Repository files navigation

cyberbullying-tweet-prediction-app

Web app Link: https://apurvayadav-cyberbullying-tweet-recognition-app-webapp-z6y643.streamlitapp.com/

This app predicts the nature of the tweet into 6 Categories.

  • Age
  • Ethnicity
  • Gender
  • Religion
  • Other Cyberbullying
  • Not Cyberbullying

Tools and Technologies

Model: Used linear Support Vector Machine to classify tweets.
Streamlit: Used Streamlit to create simple webapp to make the model interactive.

Methodology

  • Downloaded the data from kaggle. (data)
  • Performed some Exploratory Data Analysis to get the overview of data. (initial_modelling.ipynb)
  • Created a Word Cloud from the data.
  • Performed the necessary steps for textual analysis.
    • Removing Stopwords, puctuations, URLs, etc
    • Performed Stemming and Lemmatization.
  • Automated the process of preprocessing by creating functions. Which would be helpful in predicting Custom Outputs.
  • Created Illustrations for the webapp from Canva. (Images)
  • Deployed the webapp on streamlit.

References

About

This app predicts the nature of the tweet into 6 Categories. i.e. Age, Ethnicity, Gender, Religion, Other Cyberbullying, Not Cyberbullying.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published