MyKIP

Bangkit 2023 Capstone Project

Team ID : C23-PC621

ID	Name	Path
M308DSX1890	Gede Wahyu Purnama	Machine Learning
M132DSX2849	Daffa Hafiizh Permadi	Machine Learning
M257DSX0044	Daffa Arifadilah	Machine Learning
C040DSX1623	Muhammad Nabil Azizi	Cloud Computing
C304DSX1959	Arya Nur Hidayat	Cloud Computing
A304DSX1262	Muhammad Fahrizal	Android Development

Theme : Education, Learning, and Personal Development

About Project

This is the final project of Bangkit 2023, regarding the prediction of high school students who want to register themselves for the KIP (Kartu Pintar Indonesia) Scholarship provided by the government. built in the form of an android application with features: detecting students with what economic level are eligible for the KIP (Kartu Pintar Indonesia) Scholarship.

Tech Stack

Machine Learning

Datasets

Sample student data :
We got our sample students dataset from this article.

Synthesis student data :
We synthetize student data for training dataset for our model using mostly ai.

House Images :

• Google Images (web scraping using Google-Image-Scraper)
• Screenshots from Google Street View
• Own Picture
  
  

KIP Classification Model

---

Cloud Computing

Tech Stack

• Node.js
• Flask
• My SQL
• Cloud SQL
• Cloud Storage
• App Engine
• Cloud Functions

Cloud Infrastructure

API and Deployment

1. Create Google Cloud Platform Project
2. Create a App Engine, Cloud SQL My SQL, 2 Cloud Storage (FotoRumah and MLModel)
3. Create a service account in order to access Cloud SQL and Cloud Storage ( Please note the access given )

API Database

4. Build API for Database with Node JS in Google Cloud Shell (Method POST,GET,DELETE)
5. Integrate API for Database to Cloud SQL My SQL and Cloud Storage
6. Deploy To App Engine (gcloud app deploy on google cloud shell)

API Machine Learning

7. Put the model that has been created by the machine learning team into google storage (MLModel)
8. Build API Machine Learning with Flask in inline code editor Cloud Functions (Method POST)
9. Change Entrypoint same in the main.py
10. Make sure requirements.txt is the same as the enviroment used in main.py
11. Deploy to Cloud Functions (click "Deploy")

Testing API

12. Using Postman to test a deployed api
13. For API Database Use URL in App Engine (example: https://[Project_ID].et.r.appspot.com)
14. Change Method to POST to create new Students Data to Database
15. Change to Body and to raw and change the Request type using JSON
16. Create the required JSON
17. Then click SEND
18. For API Machine Learning Use Trigger in Cloud Functions (example: https://[region]-[Project_ID].cloudfunctions.net/[entrypoint])
19. Change to Body and to raw and change the Request type using JSON
20. Create the required JSON
21. Then click SEND

Build with Love by Muhammad Nabil Azizi and Arya Nur Hidayat as a Cloud Team ❤