Inspiration

A friend who was a TA would constantly complain about receiving too much email. Whilst wanting to respond to each student who had reached out, it was difficult to do so due to the sheer volume of email received. However, we observed that, at any given time of the semester, a lot of the questions being asked were essentially the same.

At the beginning of the semester, a lot of emails received would be about course pre-requisites, course registration, wait-lists, and so on. Almost each week, a lot of email would concern some particularly difficult problem on the homework for the week.

What it does

Using machine learning / natural language processing algorithms, MailEfficient cluster similar emails together. The user can, at a glance, see the top terms common to each group of similar emails, and can even skim through each email if she/he pleases. The user can then type a single response for any given cluster of similar emails, and very conveniently have it sent to each email address in that particular cluster.

How I built it

This project was built using Python, NLTK, sci-kit learn, HTML5, and the Django framework.

Built With

Share this project:

Updates