BrainTrain
A cognititive web based platform which learns from user responses to a auto generated quiz Set. The Algorithm learns the strong and weak areas of a student and helps him revise the topics timely.
Created on 10th April 2021
•
BrainTrain
A cognititive web based platform which learns from user responses to a auto generated quiz Set. The Algorithm learns the strong and weak areas of a student and helps him revise the topics timely.
The problem BrainTrain solves
It solves the problem of Short Term memory of the people which is too common to see in present generation. We can't remember things but this algorithm is fine tuned in such amanner that it allows us to cope up with the short term memory and signs of forgetting easily. It helps students in cramming the topics which are made to be crammed like General Knowledge , History , Social Sciences and BioMedical topics. So Lets start mugging up or cramming in a new fashion with Brain Train.
On the other hand it also solves the task of making quiz and question test set for students for non conceptual and theiry subjects with the help of Natural Language Processing. We combined the existing power of NLP and SPace Repetiition to suit it the best to avoid short term memory problems of the students
Methods Used
Question generator from text . (Implemented) (It generates questions and feed wrong/right answers given by a user topic wise to the backend)
Space Repetition Algorithm our Modification based Quiz (Implemented ) ( Works upon users response to figure out theor strong and weak topica and help them learn the things smoothly)
Automatic Extractive Summarizer/Notes Maker (partially implemented)
Space repetition.
Although the principle is useful in many contexts, spaced repetition is commonly applied in contexts in which a learner must acquire many items and retain them indefinitely in memory. It is, therefore, well suited for the problem of vocabulary acquisition in the course of second-language learning. A number of spaced repetition software programs have been developed to aid the learning process. It is also possible to perform spaced repetition with flash cards using the Leitner system.
Doc LInk https://docs.google.com/document/d/1RlrjunqtT8I4IgVxz02Z-slJbpAuiGNHQ6UYSja3p9E/edit#
Video Backend https://drive.google.com/file/d/104fHk4gcxZjxryIhXkMMzRoq4-nGo2iJ/view?usp=sharing
Api http://ec2-3-87-102-48.compute-1.amazonaws.com:3000/text?topics=Kinematics,Physics
Challenges we ran into
Some Technical Challenges:
Identifying the learning curve of every other person was very difficult. As everyone learns differently so identifying the algorithm adn formula which suits the best amount of people in the society is a major challenge and which needs to be solved. Though we are not guaranteed of our formulated results but this can be improved in future with collaboration from open source community.
Learning about Space repetition and implementing it in a short frame of time.
Implementing a lot of Apis , a cluster of python apis and node apis and integrating them to suit best to our needs.
Making a smooth top notch front end in a limited time frame.
Deploying Apis over Amazon ec2 and s3 within free tier and learning it was both fun and challenging.
We ran into a lot of bugs specifically on how to link the two Apis altogether within a local system it was becoming messy this we deployed out 2 servers separately and in a distributed fashion on ec2 which were integrating among themselves and giving us the one combined response. which took a lot of hard work to implement but helped us in making things faster thus we converted the challenge into a feature to support us.
Non Technical challenges
Collaborating with friends from Home and in a remote environment is not that easy. But we managed it.
Some communication gap was experienced throughout the hackathon as we were not in constant touch but eventually we fixed it soon by meeting on a hourly basis. to help each other resolve their bugs.