Socho Bot

Q&A Bots for Education, a demonstration of real-time machine comprehension and information retrieval.

What it is: Web App with chat UI
Technologies and Frameworks: Python back-end (TensorFlow, Flask)
jQuery front-end
Teammates: @Nikhilaravi
:
ABOUT THE PROJECT

SochoBot ingests text (e.g. from a wikipedia article, a blog post, a textbook, or from real-time user input) and computes a machine-representation of the text. When a student asks a question about that text, a variable-length response is retrieved from this machine representation and returned to the user.

SEE IT IN ACTION
WHY DO WE NEED AI TUTORS?

Bringing 1-on-1 learning experiences to anybody, anytime, anywhere.

Thirty years ago, educational psychologist Benjamin Bloom published his famous ‘2-sigma problem’ research, citing that students receiving 1-on-1 or small group tutoring perform 2 standard deviations above their peers, across multiple subjects and age groups. More recently, experiments have shown that adding 1-on-1 mentorship to e-learning platforms (e.g. MOOCs) provides much needed interactivity and improves completion rates and learning outcomes. Case in point: Udacity saw a 3x improvement in course completion in the presence of 1-on-1 human coaches accessible via chatbot.

The obvious challenge with 1-on-1 mentorship is cost and scale. But what if we could build artificially intelligent tutors that answer students’ questions and share relevant knowledge instantly, accurately, with little or no human intervention. What if we could do this at low cost, for everybody, anytime, anywhere.

SochoBot is an attempt to do exactly that.