🏆 Coin Prediction King - Backend

Coin Prediction King is a real-time cryptocurrency prediction game where players can select coins and compete based on market performance. This repository contains the backend infrastructure that powers the game.

📌 Project Overview

The backend architecture is designed for real-time data processing, user interactions, and deep learning analysis. The system integrates Kafka, AWS Lambda, DynamoDB, S3, and Spark for efficient data handling.

🛠️ Tech Stack

• Data Collection & Processing

  • Upbit WebSocket API
  • Apache Kafka (Producer & Consumer)
  • AWS DynamoDB
  • AWS S3 (Data Lake)

• Backend Services

  • AWS Lambda (Serverless execution)
  • AWS API Gateway (API Management)

• Deep Learning & Analytics

  • LSTM-based Prediction Model
  • Amazon Kinesis for User Data Stream
  • Spark DataFrame Analysis

• Frontend & Deployment

  • React.js
  • Vercel (Frontend Hosting)

---

📊 System Architecture

---

🔥 API Endpoints

📌 Real-Time Crypto Data

Method	URL	Request	Response
GET	/upbit_data	-	[{"trade_price": 82635000.0, "code": "KRW-BTC", "change": "RISE", ...}]

📌 User Data Management

Method	URL	Request	Response
POST	/set_user_data	{ "student_id": "20201515", "name": "홍길동", "department": "컴퓨터공학과", "nickname": "코인마스터", ... }	{ "message": "Data saved successfully" }
GET	/get_user_data	-	{ "message": "유저 데이터 가져옴", "data": [...] }
GET	/get_the_best_user_data	-	{ "message": "1등 유저 데이터 가져옴", "data": { "nickname": "찬영", "balance": 4771917000, ... } }
GET	/get_department_list	-	{ "message": "학과 리스트 가져옴", "departments": ["컴퓨터공학과", "정보보호학과", ...] }

📌 Deep Learning & Analytics

Method	URL	Request	Response
POST	/set_deeplearning	[ { "code": "KRW-BTC", "fastest_growth": true, ... }, ... ]	{ "message": "Deep learning results saved successfully" }
GET	/get_deeplearning	-	[ { "code": "KRW-BTC", "fastest_growth": true, ... }, ... ]

📌 Clickstream Data

Method	URL	Request	Response
POST	/set_click_stream	{ "user_name": "CryptoWarrior", "selected_coins": ["BTC", "ETH", "SOL"], ... }	{ "message": "Clickstream data saved successfully" }

---