Easy Syllabus

Easy Syllabus, helping students stay on track

Comment

Inspiration

We were inspired by COVID-19 and the adaptation of digital learning among youths. Leaning becomes difficult during these challenging times when we are separated from our peers and removed from classrooms. With everything being virtual, deadlines and deliverables are now harder to follow.

Therefore, we created an innovative solution that can automatically generate an ICS file you can subscribe to with your preferred calendar service from scanning your course syllabuses. We hope to ease the anxiety in missing deadlines and examinations among students during a time like now.

What it does

Easy Syllabus is a web-based service that takes a pdf/image of your course syllabus and ultimately converts it into an ICS format. The website scans the document and extracts information about the deliverable which is then displayed to the user. The user can verify the data or make changes to correct the information before exporting it as an ICS file.

How we built it

We utilized AWS Textract to assist with image recognition. Then we exported the raw data from AWS to process it in the server-less back-end of our website in python code. After we reconstruct the text, it is displayed via forms on the website that users can interact with. Finally, the data from the forms are converted into an ICS file available for download.

Challenges we ran into

Getting familiar with the AWS Textract output format. Processing the data was difficult since it was not a familiar format for us to use. Though the image recognition was mostly accurate, sometimes there can be mistakes in the outputted information.

What we learned

We learned a lot about how to extensively utilize the AWS platform for services such as Textract and getting it to work with our own site. We also learned about deploying websites with server-less back-ends. Coding in python for the back-end was also a new experience for some of our members.

What's next for Easy Syllabus

We hope to do more with Syllabus than just providing students with a calendar file. In the future, we can see Syllabus going far beyond its current limits. By using techniques in ML, we can offer custom study suggestions for the semester from simply scanning course syllabuses. Through training a model, we hope to provide a predictor that can offer a personalized study plan based on "course hours, assignment weightings, assignment due dates, etc."

Built With

Share this project:

Updates