Skip to content

Latest commit

 

History

History
198 lines (158 loc) · 7.03 KB

README.md

File metadata and controls

198 lines (158 loc) · 7.03 KB
Raw

Spring Boot Chatbot Project

Welcome to the Spring Boot Chatbot project! This application integrates multiple technologies and libraries to deliver a robust chatbot experience, including SMS interactions, PDF document reading, and natural language processing capabilities. Below is a detailed description of the project, including setup instructions, dependencies, and key features.

Table of Contents

Prerequisites

  • Java 17: Ensure that Java Development Kit (JDK) version 17 is installed on your machine.
  • Maven: This project uses Maven for build and dependency management. Install Maven from here.
  • Neo4j: Download and set up a Neo4j database from here.
  • OpenAI API Key: Register and obtain an API key from OpenAI.
  • Africastalking API Key: Register and obtain your username and API key from Africastalking.

Setup Instructions

  1. Clone the repository:

    git clone https://github.com/Titus-Nyandoro/Chatbot.git
cd Chatbot

  2. Build the project:

    mvn clean install

  3. Run the application:

    mvn spring-boot:run

Configuration

  1. OpenAI API Key:

    • Add your OpenAI API key to the application.properties file:
    openai.api.key=your_openai_api_key

  2. Africastalking Configuration:

    • Add your Africastalking username and API key to the application.properties file:
    africastalking.username=your_username
africastalking.api.key=your_api_key
africastalking.shortcode=your_sms_shortcode

  3. Neo4j Database URL:

    • Add your Neo4j database connection details to the application.properties file:
    neo4j.url=bolt://localhost:7687
neo4j.username=your_neo4j_username
neo4j.password=your_neo4j_password

  4. System Message Configuration:

    • Define the system message by modifying the system-message.txt file located in the src/main/resources folder. This message will be used to guide the chatbot's responses.

  5. Load PDF Documents:

    • Add important PDF documents to the docs folder located in src/main/resources. These documents will be parsed by the chatbot for information retrieval.

Project Structure

The project follows a typical Spring Boot structure:

spring-boot-chatbot/
│
├── src/
│   ├── main/
│   │   ├── java/com/example/chatbot/
│   │   │   ├── config/      # Initial Configs
│   │   │   ├── controller/  # Controllers for handling API requests
│   │   │   ├── service/     # Service layer for business logic
│   │   │   ├── model/       # Domain models and DTOs
│   │   │   ├── repo/        # repositories
│   │   │   ├── service/     # Services for API processing
│   │   │   ├── utils/       # Necessary app utils
│   │   │   └── ChatbotApplication.java  # Main Spring Boot application class
│   │   └── resources/
│   │       ├── application.properties  # Configuration file
│   │       ├── system-message.txt  # System message for the chatbot
│   │       ├── docs/  # Folder for PDF documents
│   └── test/  # Unit and integration tests
│
└── pom.xml  # Maven build file

Key Features

  • PDF Document Reader: Integrates with Spring AI PDF Document Reader for document parsing.
  • Neo4j for Vector Database: Utilizes Neo4j for efficient storage and retrieval of vector data.
  • OpenAI Integration: Uses OpenAI for text embedding and completion.
  • SMS Interaction: Implements SMS parsing and sending using the Africastalking library with a response limit of 144 characters.

API Endpoints

Send SMS

  • Endpoint: /send
  • Method: POST
  • Request Body: SmsRequest
  • Response: SmsResponse
@PostMapping("/send")
public ResponseEntity<SmsResponse> SendSms(@RequestBody SmsRequest smsRequest) {
    log.warn("Send sms {}", smsRequest);
    return ResponseEntity.ok(smsService.sendSms(smsRequest));
}

Receive SMS

  • Endpoint: /incoming
  • Method: POST
  • Consumes: application/x-www-form-urlencoded
  • Request Body: IncomingSmsDTO
@PostMapping(value = "/incoming", consumes = "application/x-www-form-urlencoded")
public void processIncomingSms(@ModelAttribute IncomingSmsDTO smsPayload) {
    log.warn("Incoming sms {}", smsPayload);
    smsService.processIncomingSms(smsPayload);
}

Dependencies

  • Spring Boot: Core framework for building the application.
  • Maven: For project build and dependency management.
  • Spring AI PDF Document Reader: To handle PDF document reading.
  • Neo4j: For vector database functionality.
  • OpenAI: For embedding and text completion capabilities.
  • Africastalking: For SMS functionalities limited to 144 characters.

Key Dependencies in pom.xml

<dependencies>
    <!-- Spring Boot Starter -->
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter</artifactId>
    </dependency>
    
    <!-- Neo4j Dependency -->
    <dependency>
        <groupId>org.neo4j.driver</groupId>
        <artifactId>neo4j-java-driver</artifactId>
    </dependency>
    
    <!-- OpenAI Dependency -->
    <dependency>
        <groupId>com.openai</groupId>
        <artifactId>openai-java-client</artifactId>
    </dependency>
    
    <!-- Africastalking Library -->
    <dependency>
        <groupId>com.africastalking</groupId>
        <artifactId>africastalking-sms</artifactId>
    </dependency>
</dependencies>

Usage

  1. Sending SMS:

    • Send a POST request to /send with SmsRequest payload to send an SMS.
    • The system will respond with SmsResponse.

  2. Receiving SMS:

    • Configure your SMS gateway to forward messages to /incoming.
    • The system will process incoming messages automatically.

  3. Loading PDF Documents:

    • Ensure that the docs folder under src/main/resources is loaded with PDF files that contain important information. The application loops through these files to parse and use the content.

  4. Configuring System Message:

    • Modify the system-message.txt file located in src/main/resources to set up a customized system message for the chatbot.

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any enhancements or bug fixes. Ensure your code adheres to the project’s coding standards and includes relevant tests.

License

This project is licensed under the MIT License - see the LICENSE file for details.