By Emily Glassberg Sands, VP of Data Science at Coursera, and Marianne Sorba, Data Scientist at Coursera

Since we launched the Coronavirus Response Initiative on March 12, more than 2,600  colleges and universities around the world have activated Coursera for Campus programs to take learning online and minimize student disruption. We’re humbled by the global response and are working hard to be even more useful to universities who need to move online quickly. 

As universities go live using our offering, they urgently need an easy solution to help identify courses on Coursera that most closely match each course in their on-campus catalogues. Manual curation is too slow when it’s to be done across thousands of universities and millions of on-campus courses, especially when faculty and staff are already stretched thin. Two weeks ago, the Data Science team at Coursera started developing a natural language processing solution to automate the matching and minimize the need for human curation.  

Today, we are announcing CourseMatch, a machine learning solution that ingests a school’s on-campus course catalogue and matches each course to the most relevant courses in Coursera’s catalogue of 3,800 courses. It can ingest catalogues in more than 100 languages and map them to the most relevant courses in English or in any of the 50+ languages (translated and subtitled) available on Coursera. This enables universities in the US and internationally to quickly deliver relevant courses to their students.  

The solution has already matched more than 2.6 million on-campus courses across 1,800 schools to courses on Coursera — from Johnson C. Smith University in the US to Universidad Autonoma de Barcelona in Spain to the AICTE curricula in India. 

Here are a few examples of mapping for institutions based on publicly available information:

CourseMatch returns up to five most relevant courses on Coursera for each course in the on-campus catalogue. It also returns the relevance score, which is normalized within college or university with higher scores representing stronger matches.

We are excited to see that the solution is already helping schools respond to students’ needs. According to Terik Tidwell, managing director of the Smith Tech Innovation Center, Johnson C Smith University, “Coursera is helping us to quickly take our programs virtual while maintaining program integrity and quality.”

Colleges and universities can access the solution directly at CourseMatch and search for their institution’s catalogue. If your catalogue is not yet among the 1,800 available, email a CSV of your offerings to coursematch@coursera.org and your catalogue will be mapped and included on CourseMatch within 2 business days. 

Our goal is to help every college and university in the world move online as quickly and easily as possible. This is our first iteration of CourseMatch, and we look forward to refining our machine learning models based on your feedback. We invite you to try this solution in your efforts to quickly scale programs online during the COVID-19 crisis. 

The post Coursera launches CourseMatch: A machine learning solution that automatically matches a University’s on-campus courses to courses on Coursera appeared first on Coursera Blog.