Inspiration
Reading articles and other forms of literature may seem overwhelming, as the style itself is dense and uninteractive in most cases. I wanted to see if I could create a starting point from an algorithm to provide some more interactivity and knowledge.
What it does
The program asks for a pdf file you would like to "glossarize" and how many words wanted. Then it would extract all unique words from the article, filter it, and extracts the number of "complex" words. In this case, I defined "complexity" as how infrequent it is. Then, the program extracts the meanings of each word and displays them.
How I built it
Python was the main backend for this. More specifically, the nltk package helped a lot with the filtering and organizing by word frequency. To get the definitions, I used a dictionary API to match the words to their definitions. Finally, I deployed it on a web application using Flask.
Challenges I ran into
The biggest challenge I ran into was the front-end development, as I was very unfamiliar with that territory. I believe I managed to get the suitable outcome I desired, but further improvements are necessary.
Accomplishments that I'm proud of
I am very proud of my understanding and utilization of a new Python package.
What I learned
I learned more about the nltk package in Python, how to extract data using an API, and finally the general process for using Flask
What's next for Glossarize
As mentioned earlier, the display for Glossarize could be cleaned up a lot better and displayed in a better format. I could also add a highlight tool, where I display the pdf, and highlight the words I was defining.
Log in or sign up for Devpost to join the conversation.