This project is a Chat Support Service built using Java and Spring Boot. It provides various functionalities to manage and analyze chat interactions between agents and customers. The main features include:
-
User Management:
- Retrieve user scores by their ID.
- Retrieve lists of agents and customers with their scores.
- Retrieve the best and worst performing agents based on their scores.
- Retrieve agents and customers sorted by their scores.
-
Chat Management:
- Save chat interactions and calculate sentiment scores for the chats.
- Update the scores of agents and customers based on chat sentiment analysis.
-
Sentiment Analysis:
- Calculate sentiment scores for chat messages using an external sentiment analysis service.
- GET /api/v1/user/health: Check if the User Service is running.
- GET /api/v1/user/agent-score: Get the score of a specific agent.
- GET /api/v1/user/customer-score: Get the score of a specific customer.
- GET /api/v1/user/agent-scores: Get scores of all agents.
- GET /api/v1/user/customer-scores: Get scores of all customers.
- GET /api/v1/user/best-agent: Get the agent with the highest score.
- GET /api/v1/user/worst-agent: Get the agent with the lowest score.
- GET /api/v1/user/agents-by-score: Get agents sorted by their scores.
- POST /api/v1/chat/save: Save a chat and calculate its sentiment score.
- Java
- Spring Boot
- Maven
- MongoDB
- Clone the repository.
- Navigate to the project directory.
- Run
mvn clean installto build the project. - Run
mvn spring-boot:runto start the application.
- Configure the MongoDB connection and external sentiment analysis service URL in the
application.propertiesfile.
This service uses a pre-trained BERT model for sentiment analysis via a Flask API provided by the Chat Support Sentiment Analysis project. Ensure that the Flask service is running and accessible to this application for sentiment analysis functionality.
This project is licensed under the MIT License.