Educational Data Science
Our solutions are powered by machine learning methods for early identification of students at risk of failing
We Help
EdTech Startups
We are tech-enablers. We turn bright ideas into actual products.
Academic Institutions
We help universities, colleges, and schools deliver even better educational services.
Nonprofit organizations
From course providers to free libraries, we build EdTech solutions for non-profit organizations of any type.
Our Solutions
eLearning and Education Management Platforms
Student Assessment and prediction
Educational Reporting and Analytics Systems
Prediction Model of at-Risk Students
Comprehensive Analytical Methods
Relational Data to Actionable Data
Transforming the student engagement information, through smart data processing, to extract concrete student activity information, attendance, assignment score, quizzes, class participation etc.
Feature Reduction & Model Optimization
The model further runs series of optimization techniques and feature reduction models. Reduces the features through a number of feature reduction techniques, such as SHAPLY, Extra Tree classifiers, PCA, etc.
Optimized Deep Prediction Model
We utilize the extracted week-wise information and feed into its own optimized Long Short Term Memory(LSTM) model for the early prediction of students at-risk. Students at-risk can be warned by sending alert.
Platform for Reporting and taking Prompt Actions
92% Accuracy in the prediction of dropouts only by mining up to 12th-week data of students engagement on VLE. Students are likely to drop out from a course, with an accuracy scale ranging from 78.16% in the 1st week.
