Mind your map

Hate making notes for your classes? No worries! Mind your map, an application that'll convert your transcript into mindmaps, FOR YOU! Doesn't matter if it's a book or transcript, we got you covered

Comment

Inspiration

As students of zoom university, we first hand understand the pain of visualizing tonnes of information from the lectures and textbooks. To stop this uncountable amount of rewatching lectures, we created Map your mind to convert text into mind-maps that summarize key concepts and graphically show their connections.

What it does

Map your mind takes any form of text input like your zoom lecture transcript, a chapter of your book and then finds the most import keywords and parent relationships to create mind-maps. So, a student simply uploads their zoom lecture transcript and we get the most important keywords and topics out of it to return an interactive mindmap for them!

How we built it

Front End: We created a webpage using html, css & javascript which acts as the interface between the user's input and the python script for getting the keywords

  1. Used REST api from pyflask to connect the javascript to python
  2. Implented the algorithm to find the keywords for given text using natural language processing. Used gensim, spacy libraries and features like lemmatization. We aim to use an LDA model to get important topics of the text and keyBERT model to extract important keywords for these topics.
  3. We use basic graph algorithm to construct the parent nodes and leaves.
  4. We a GO JS plugin to visualize the app. ## Challenges we ran into connecting python with javascript, fetch api, creating the most efficient algorithms. Getting a video for the demo ## Accomplishments that we're proud of We achieved our MVP and actually created a fully fledged working algorithm and webpage. ## What we learned natural language processing, rest api, web development, design ## What's next for Notes2MindMap! Achieve database management with the option for the user to save their mindmaps for different chapters. Complete all the functionalities

Built With

Share this project:

Updates