Skip to content

README.md

Latest commit

 

History

History
34 lines (26 loc) · 1.7 KB

README.md

File metadata and controls

34 lines (26 loc) · 1.7 KB

KeebsChat

Mechanical Keyboard Chatbot, designed to enhance the keyboard community and help new comers learn more about this awesome hobby and promote local keyboard businesses in Canada!

How It Was Built:

This program was made with the programming language Python along with a few frameworks such as Tensorflow for creating and training neural networks, NLTK for tokenizing & stemming words, Flask to provide a webpage for the chatbot and numpy to manage data. The full list of languages & frameworks are listed below:

  • Python
  • Tensorflow
  • Tflearn
  • Flask
  • NLTK
  • Numpy
  • Json
  • Pickle
  • Beautifulsoup (bs4)
  • HTML + CSS

KeebsChat In Action:

Basic_functionality.mp4
Complex_functionality.mp4

Future Improvements:

Some of the future improvements/features I would add to this program are listed below:

  • Make a .exe file for users to directly download and use the program without to much setup needed.
  • Talk to local computer keyboard businesses and try to implement the chatbot into their website.

Updates:

  • 6/18/2021 - Added a python webscraper program to scrape keyboard product info from https://www.apexkeyboards.ca/. This scraper will be necessary for the chatbot's main functionality.
  • 7/09/2021 - Completed all functionality for the chatbot and fixed some errors in apexkeyboards webscraper, will be fixing up bugs and cleaning the code.
  • 7/28/2021 - Code is cleaned and bugs are fixed. Next I will implement a Flask webpage to host KeebsChat.
  • 8/28/2021 - I have finished the Flask webpage for the chatbot and now have completed this project. Yay!