This post is by Roopali Kaujalgi, Senior Program Manager at the Data Group at Microsoft.
If you’re a teacher or school administrator, imagine how powerful it would be to proactively identify at-risk learners, help improve their performance and drive up overall graduation rates at your educational institution. Applying data science principles to student data and creating a better foundation to collect and analyze student information can help drive improved education outcomes. Using machine learning algorithms to illustrate patterns or make predictions with high confidence levels can alert educators to insights they just didn’t have before, and even determine a better course of action.
The Cortana Intelligence Gallery is a community site where data scientists and developers can consume and share analytics solutions built using the Cortana Intelligence Suite.
Some solutions pertaining to education that are currently available in the Gallery include the following:
Predict if a student will solve a problem in their first attempt: Given a problem ID, brief description, student ID, timestamp and how many attempts the student took to solve the problem in the past, the solution predicts if a student will solve a given problem in their first attempt.
Predict a student’s performance in Mathematics: This solution compares different classifier models to predict a student’s performance in Mathematics.
Predict the likelihood of a student dropping out of a MOOC: This is a typical customer churn analysis problem, but in the context of MOOC platforms.
We’d also like to share a few interesting customer stories pertaining to the use of the Cortana Intelligence Suite in education:
- When the Cleveland Metropolitan School District wanted to change how it delivered education, it looked for a better way to predict student and school performance. The district worked with Microsoft and Neal Analytics, an independent software vendor, to build a cloud-based data visualization and predictive-analysis solution using Microsoft Power BI and Azure Machine Learning. Now the district can monitor student performance, identify learning barriers and act more quickly to help the children who need the most assistance. The net result is that the school district is able to give every student a better chance.
- In this video, you can learn how the Tacoma Washington School District used Azure Machine Learning to predict students’ risk of dropping out of school. The district has been able to use their system to improve their graduation rates from 55% to 82.6% in 6 years!
- In August 2015, as part of the “Badi Pilusthondi’ (The School is Calling) initiative launched by the chief minister of Andra Pradesh, a state in India, the government and Microsoft announced a cloud-based machine learning solution in the field of education using Microsoft Azure ML and Power BI. The pilot project was implemented in the Chitoor district. The project is aimed at analysing existing student data to predict and reduce dropouts through early detection, timely intervention and appropriate corrective measures
See how the solutions in the Cortana Intelligence Gallery may be able to help you improve student attainment and performance in your educational institution. We would love to hear your stories.