INSPIRATION
Most of our day to day conversations are unstructured. This only extends to from our simple chat conversions to articles, newspaper, emails and much more. Conversations seems to be all over the place and in this chaos it hard to lose sight of the most insightful information hidden within the body of text.
WHAT IT DOES
With the use of NLP, Natural Language Processing, the text is classified, and the classifier automatically analyze it and puts it in an organized structure with all the important details.
BUILDING PROCESS
The project was divided into two major component. First, development and deployment of libraries in python. Second, creating the application to get the input and represent it to the user.
Stage 1) Created the API key for Monkey Learn
Stage 2) Created the classifiers on Monkey Learn
Stage 3) Developing the python libraries for extraction and classification
Stage 4) Creating the desktop application using Bootstrap, HTML/CSS/JS
Stage 5) Creating the Eel Libraries to bridge the application with the main Engine
CHALLENGES
The major challenge that we faced was collecting data to train the classifiers. It was challenging because we had to make sure and avoid all the basis in our data so the classifier isn't biased.
ACCOMPLISHMENTS
1) Successfully achieved our goal
2) Added four more types of documents that can be classified, (News, Tweets, Chats, and Emails)
3) Integrating and learning about the APIs.
4) Exploring and implementing new technologies.
NEXT STEPS
Next steps for our project is to make it robust by adding more analyzing tools and giving the option to the user to create tags and organize their text based on the classified information. In long term, we can also work on bringing the project to Android and iOs.
Log in or sign up for Devpost to join the conversation.