Inspiration
As McGill students, we often find the process of joining and finding university clubs overwhelming and confusing. The various amount of options is often not readily available or easily accessible. Email newsletters from clubs are informative about upcoming events, but manually going through each newsletter can be time-consuming and overwhelming so we wanted to automate the process.
What it does
McGill Clubs Enhanced's purpose is to give easy and quick accessibility of McGill club's calendar and information to students. The website displays a calendar that contains any upcoming events and deadlines of McGill clubs (or societies) that the student might choose based on the filters that they pick.
How we built it
This project utilized JavaScript for backend programming and implemented web scraping techniques to extract relevant information from email content. The extracted data about upcoming events from university club newsletters, was then processed through OpenAI. The output of this processing was structured into a JSON file, forming a collection of event objects. This approach automated the extraction and organization of event details, contributing to a more efficient and streamlined process for obtaining information on upcoming events from diverse sources.
Challenges we ran into
It was difficult to gain access into gmail content and working with the Gmail API from Google Cloud proved to be a difficult connection.
Accomplishments that we're proud of
We're happy to have managed a project from conception to implementation, involving tasks such as planning, coding, testing, and integrating different components.
What we learned
We learned how to integrate external AI services into our web application involving understanding APIs, making requests, and handling responses from the OpenAI service.
What's next for McGill Clubs Enhanced
We look forward to making use of more precise technology to scrape data from newsletters, social media posts (ie. through Instagram, Facebook, TikTok, etc.) to give better information about clubs to our users. More specifically, we hope to be able to use technology that could scrape data from images.
Built With
- css
- html
- javascript
- json
- openai
Log in or sign up for Devpost to join the conversation.