HealthPredict AI — Student Code Package

LKS Cloud Computing - Tingkat Provinsi 2026

Code package for the HealthPredict AI competition. Please read the module carefully before using these files.

---

File Structure

healthpredict-student/
├── glue/
│   └── healthpredict_etl.py        PySpark ETL — 9 Steps for Data Transformation
├── sagemaker/
│   ├── sagemaker_pipeline.py       Register a 3-step SageMaker Pipeline
│   ├── processing_script.py        Script Processing Step (SKLearnProcessor)
│   ├── deploy_endpoint.py          Deploy model Approved to real-time endpoint
│   └── batch_transform.py          Submit Batch Transform job to S3
├── lambda/
│   ├── predict.py                  POST /predict — validasi, normalize, invoke, save
│   ├── history.py                  GET /history and GET /predict/{patient_id}
│   └── health.py                   GET /health — check status endpoint and DynamoDB
├── athena/
│   └── athena_queries.sql          5 dataset exploration queries
├── scripts/
│   └── redshift_schema.sql         DDL: 4 tabel, 3 view, 1 stored procedure
└── frontend/
    └── index.html                  Single-page web app (deploy to AWS Amplify)

---

---

Required Configuration

sagemaker_pipeline.py and deploy_endpoint.py and batch_transform.py

Replace the following values before running:

STUDENT_NAME = "yourname"   # ← Replace with your name (lowercase, no spaces)

Lambda Environment Variables

Variable	Value
SAGEMAKER_ENDPOINT_NAME	healthpredict-endpoint-diabetes
DYNAMODB_TABLE_NAME	healthpredict-predictions
SNS_TOPIC_ARN	Your ARN SNS topic
REDSHIFT_SECRET_ARN	Your ARN Secrets Manager secret
RISK_THRESHOLD_HIGH	0.7
RISK_THRESHOLD_LOW	0.3
PREDICTIONS_BUCKET	healthpredict-predictions-[nama]-2026

---

Important Note

• All resource names must be prefixed healthpredict-
• S3 bucket: healthpredict-[service]-[namaAnda]-2026
• IAM Role: use LabRole for all service
• Region: us-east-1