Inspiration
Wikipedia is the go to tool for accessing articles on various subject matters. With Searchify, we hope to bridge the gap between writing and research by bringing the tools that the web provides right into the very text they type.
What it does
As the writer types, a script concurrently picks out keywords and topics related to the writer's current work. Powered by Natural Language toolkit, a python based NLP model, relevant sources are consulted and returned to the front-end. From there, articles are displayed wikipedia style, set over the writing in the form of links.
How we built it
We began by building a simple text editor for the writer to work in. It is equipped with basic features such as italicizing, bolding and font size adjustment. Then we built the python script responsible for the main feature: determining important features in the writing and finding useful sources. Then, we worked on combining the results of the Language Processing and reintegrating them in the front end.
Challenges we ran into
The hackers on our team were not very familiar with the bridge between python and typical HTML/CSS/Javascript front end. A lot of research had to be done
Accomplishments that we're proud of
We learned a lot through this experience and tackled challenges we had never seen before.
What's next for Searchify
Due to time, a lot of features could not be fleshed out as we hoped. As we continue to work on the project beyond HackMIT, we will be able to further improve our methods and fine-tune our parameters.
Log in or sign up for Devpost to join the conversation.