Cryptora-Analyze-Service is a Spring Boot application that retrieves cryptocurrency data from the Binance API and analyzes it based on price fluctuations. The service tracks price movements of various cryptocurrencies and assesses both short-term and long-term trends, utilizing the Average True Range (ATR) with configurable periods and multipliers to enhance its insights. A key advantage of this application is its ability to transmit data to other services or bots, such as Telegram, using RabbitMQ, facilitating seamless integration and communication for generating trading signals (buy, sell, hold).
This service does not guarantee a 100% increase or decrease in cryptocurrency values. The service was developed solely by the author for educational purposes and should not be considered as financial advice. The creator of this service is not responsible for any financial losses or damages incurred while using it. Cryptocurrencies are highly volatile and carry inherent risks. Users should conduct their own research and consult with a qualified financial advisor before making any investment decisions. By using this service, you acknowledge and accept these risks.
-
Real-time Data Fetching: The application integrates with the Binance API to retrieve live cryptocurrency data. Users can track any available cryptocurrencies.
-
Cryptocurrency History Storage: MongoDB is used to store the history of cryptocurrency prices, allowing users to analyze historical growth and decline trends.
-
Price Trend Analysis: With stored price data, users can perform trend analysis on cryptocurrency growth and decline, supported by the Ta4j library.
-
Trading Signal Generation and Delivery: Generated signals are sent via a RabbitMQ message queue, making it easy to connect and deliver this data to other services or end users.
-
AI: Generation of recommendations based on cryptocurrency data analysis, creating trading advice, price forecasting, and trend identification.
-
Java: The primary programming language.
-
Spring Boot: Framework used for building the service.
-
MongoDB: A NoSQL database used for storing real-time cryptocurrency quotes.
-
RabbitMQ: Message queue for asynchronous delivery of trading signals, enabling integration with other systems.
-
Ollama: A tool for local data analysis and task automation.
-
Docker: Containerization platform that helps package the application with its dependencies, ensuring consistent environments and simplifying deployment.
The application, deployed in Docker, performs cryptocurrency analysis and provides recommendations, such as "Buy", "Sell" or "Hold". The logs show the real-time analysis results for each cryptocurrency.
In the example above:
The analytic service starts the analysis and determines that for ETH, BTC, and TON, the recommended actions is to sell BTC and hold ETC and TON.
By visiting http://localhost:8088/api/v1/report/BTC, We can view a report of the current price of BTC
cryptocurrency, along with recommendations.
After successfully retrieving data from the Binance API, all cryptocurrency information is stored in the database.
In the example above:
- Amount: The quantity of cryptocurrency involved in the transaction or trading period.
- Close price: The price at which the last trade was executed during the specified period.
- Date time: The timestamp in UTC format representing when the data was recorded.
- High price: The highest price reached during the specified time period.
- Low price: The lowest price reached during the specified time period.
- Open price: The price at which the first trade occurred during the specified period.
- Ticker: The symbol representing the cryptocurrency pair being traded.
- Volume: The total amount of cryptocurrency traded during the specified period.
After analysis, recommendations are sent to a RabbitMQ queue named CryptoraQueue. RabbitMQ is used to manage messages, with each message representing a recommendation for a specific cryptocurrency. These messages can be processed by other services subscribed to the queue.
The image above shows a queue with multiple messages:
- Message 1: Recommendation to Hold ETH.
- Message 2: Recommendation to Hold BTC.
- Message 3: Recommendation to Sell TON.
Each message contains additional attributes, such as priority, delivery_mode, and content_type, which allow the system to handle them flexibly and provide recommendations in real time.
- Java 23
- Gradle 8.10.2
- Docker 27.2.0
- Binance Account
-
Clone the Repository
git clone https://github.com/dzenthai/Cryptora-Analyze-Service.git cd Cryptora-Analyze-Service -
Add Environment Variables Before setting environment variables, make sure your Binance account is verified, as this is required to obtain the API_KEY and API_SECRET.
You can obtain the API key and secret by following this link.
Create an .env file and add the required environment variables such as the Binance API key and secret.
-
Build the Project Using Gradle
./gradlew build
-
Run the Application Using Docker
docker-compose up --build
To add new cryptocurrency, update the Ticker class by adding them to the Ticker enum:
enum Ticker {
BTC,
ETH,
TON,
// add a new cryptocurrency here
}To modify the interval at which the application fetches and analyzes price data, update the following line in
the AppScheduler class:
@Scheduled(fixedRate = 10000) // Time is in milliseconds
public void executeInSequence() {
}You can adjust the short-term and long-term periods for the moving average in AnalyticService using parameters in the
application.yaml file:
cryptora:
short:
time:
period: 50 #short-term
long:
time:
period: 200 #long-term
atr:
period: 14 # atr period
multiplier: 2.0 # atr multiplierAlso, it is recommended to adjust the duration in the AnalyticService class to match, as shown below:
private Bar buildBar(ZonedDateTime endTime, Quote quote) {
return new BaseBar(
Duration.ofHours(1) // Also supports setting in seconds, minutes, days, etc.
// Other settings...
);
}