TwittReport

Page 1 of the report
Page 2 of the report
Graphical User Interface

Inspiration

We were inspired by many great web-scraping and data visualization projects from other hackathons. We have also decided to associate our project with a social media so it would resonate with more users.

What it does

It scrapes users tweets, analyzes them, and then saves the analyzed report into a nice looking pdf! (Sounds like a piece of cake, but trust me, theres a lot of things going on under the hood)

How we built it

To build TwittReport we went with Python programming language as it had all the libraries we needed for our task. It also gave us a development speed boost which was handy as we did not have much time. To accomplish our goal, we used the following libraries:

twint - To scrape tweets
spacy - To analyze tweets
tkinter - Graphical User Interface
matplotlib - Graphics for the report
numpy / pandas - Required by matplotlib
LaTeX / pdflatex - To generate a pdf

Challenges we ran into

Limited time - we were in a rush the whole process, but thankfully manage to accomplish everything we wanted.
Libraries - using many different libraries is confusing and requires a lot of work to use them properly.
Look-and-feel - we tried hard to make the pdf and gui look neat, as close to real-life products as possible.

Accomplishments that we're proud of

Flawless retrieval of tweets
Fantastic language analysis
Fenomenal graphics generation
Fabulous looking pdf
Flamboyant GUI

What we learned

We learned how to use various python libraries such as matplotlib, spacy, twint and tkinter. We learned how to manage our project under time limit and how to plan short-term projects. We also learned how to put those libraries all together to make a meaningful and useful program.