SAGE

Have you ever asked your advisor a question only to wait weeks for a response? Many times, advisors get flooded with questions that have simple answers or problems that are easy to resolve. SAGE is an advisor chatbot that can answer your questions quickly and efficiently, so you don’t have to worry about it. On top of this, SAGE can view your profile to help you make decisions about future classes and even generate personalized degree plans. Take the stress out of the logistics and just ask SAGE for help with your inquiries and classes!

MVP (Minimum Viable Product)

Create and train a chatbot that can answer FAQs listed on the ECS Advising Website using Amazon Lex and Amazon Kendra
Create an algorithm that can generate a degree plan based on major, credit hours, classes taken
Import Transcript in and use OCR/Parser to get classes, major, GPA, credit hours automatically
Implement User Authentication and adapt for a multitude of majors
Advisor Side Interface(includes appointments and priority based assistance)

Stretch Goals

Make the bot multilingual to remove language barriers
Expand the scope from just advising to everything UTD-related
Add the ability to connect to other students that would be able to help when the bot does not have an answer to a question
Utilize Nebula API and or look at things like UTD Grades Data and RMP Data(Could take a look at sk.edge architecture)
Let users manually add completed credits to the degree plan (so that it will generate future classes assuming that credit is already successfully taken)
v1.0 at Presentation Night

Milestones

Assign roles - Who will work on frontend & backend?
Set up development environments
- IDE & Necessary Plugins
- Frameworks
- Git
Learn about the stack
- Youtube tutorials
- Create demo applications
- Guides and articles about the languages & technologies
Frontend
- Brainstorm and create wireframes for the UI/UX
- Create a prototype for the final website design or app pages
- Develop the pages and continuously work with the backend to ensure each component is integrated:
  - Chatbot UI
  - Degree Plan Generation page
Backend
- Design the backend workflow
  - Determine which APIs will be used
  - Outline the cloud infrastructure (databases, storage, etc.)
- Build each backend component
  - Create a database to store important information
  - User authentication API
  - Chatbot
  - Degree Plan API
  - UTD Grades/RMP for Teachers
Final integration of all features to connect and test
Prepare for the presentation

Tech Stack

Wireframing: Figma
IDE: VSC
Frontend: React Or React Native with Expo
- React is an industry applicable library utilized for frontend with web applications.
- React Native is a cross compatible framework that paired with expo allows you to bring your apps to life while in development phases
Backend: Amazon Kendra and Lex, DynamoDB/S3 Or Firebase/Firestore, Amazon Textract or Microsoft Azure Computer Vision OR Google Cloud Vision API.
- Kendra and Lex are for training and natural language processing
- DynamoDB and S3 continue with the fully AWS approach however Firebase/Firestore work great as needed
- Amazon Textract/Azure CV/Google Cloud Vision for OCR of Transcripts

Software to Install

Tutorials and Resources

General

Front-end

Back-end

Git Commands

Command	What it does
git branch	lists all the branches
git branch "branch name"	makes a new branch
git checkout "branch name"	switches to speicified branch
git checkout -b "branch name"	combines the previous 2 commands
git add .	finds all changed files
git commit -m "Testing123"	commit with a message
git push origin "branch"	push to branch
git pull origin "branch"	pull updates from a specific branch

Meet the Team

Frontend
- Shraddha Subash
- Maruthi Eranki
Backend
- Ethan Varghese
- Ram Gudur
Project Manager: Abis Naqvi
Industry Mentor: Suraj Khosla

acm-projects/Sage