Skip to content

Chatbot_NLP_Demo.md

Latest commit

 

History

History
142 lines (107 loc) · 4.71 KB

Chatbot_NLP_Demo.md

File metadata and controls

142 lines (107 loc) · 4.71 KB

RAG+NLP Chatbot Demo for Test Automation Framework

🏃‍♂️ Execution Instructions

  1. Build the project: 
    mvn clean package
  2. Run the Chatbot GUI:
    • From the command line: 
      java -cp target/classes org.k11techlab.framework.ai.chatbot.ChatbotGUI
    • Or, from your IDE, run the main method in ChatbotGUI.java.
  3. Run RAGKnowledgeChatbot or StrictFrameworkRAGChatbot from CLI: 
    java -cp target/classes org.k11techlab.framework.ai.chatbot.RAGKnowledgeChatbot
java -cp target/classes org.k11techlab.framework.ai.chatbot.StrictFrameworkRAGChatbot
    • Or, create a small demo class with a main method that instantiates and uses these classes.
  4. Configure AI providers and RAG settings:
    • Edit config/chatbot.ai.properties to set provider order, max tokens, context length, etc.
  5. (Optional) Use screenshots and code samples below to explore features.

🚀 Overview

This demo showcases the Retrieval-Augmented Generation (RAG) and NLP-powered chatbot built into the K11 Tech Lab Selenium Java Automation Framework. The chatbot delivers context-aware, framework-specific answers, generates test cases, and provides conversational assistance for automation engineers.

Key Features

  • Retrieval-Augmented Generation (RAG): Combines LLMs (OpenAI, Ollama, HuggingFace, LM Studio) with a persistent knowledge base.
  • Configurable AI Provider Pipeline: Select and prioritize providers via config.
  • Strict Knowledge-Base-Only Mode: Ensures answers are grounded in your framework docs/code.
  • Modern Chatbot GUI: Branded, user-friendly interface with source references and timestamps.
  • Test Automation Ready: Generates, explains, and troubleshoots Selenium test cases.

1. Launching the Chatbot GUI

Chatbot GUI Screenshot

Run the following command or launch from your IDE:

public static void main(String[] args) {
    javax.swing.SwingUtilities.invokeLater(() -> {
        new ChatbotGUI().setVisible(true);
    });
}

2. Asking Framework-Specific Questions

Framework Q&A Screenshot

String question = "How do I implement a self-healing Selenium locator?";
String answer = chatbot.answerFrameworkQuestionOnly(question);
System.out.println(answer);

3. Generating Test Cases

Test Case Generation Screenshot

String testPrompt = "Generate test cases for an e-commerce checkout flow.";
String testCases = chatbot.answerFrameworkQuestionOnly(testPrompt);
System.out.println(testCases);

4. Configuration Example

# chatbot.ai.properties
chatbot.rag.maxContextLength=3000
chatbot.rag.maxTokens=1200
chatbot.rag.maxRetrievedDocs=3
ai.provider.priority=OPENAI,OLLAMA,SIMPLE

5. Why Use This Chatbot?

  • Get instant, accurate answers grounded in your own framework.
  • Save time, reduce onboarding friction, and boost test coverage and quality.
  • Robust, extensible, and ready for enterprise use.

6.🗺️ Chatbot RAG+NLP Flow (Mermaid)

flowchart TD
      %% User Layer
      A[User / Engineer] -->|Natural Language Query| B(Chatbot / NLP Engine)

      %% NLP & Intent
      B -->|Intent Recognition| C{Intent}

      %% Intent Branches
      C -->|Knowledge Query| D[RAG Engine]
      C -->|Framework Action| M[Automation Framework]

      %% RAG Internal Flows
      D -->|Semantic Search| E[Knowledge Base]
      D -->|Embedding Lookup| F[Embedding Cache]
      D -->|LLM Completion| G[LLM Provider Manager]

      %% Providers
      G --> H1["Cloud / Local LLMs"]
      H1 --> H2["OpenAI, HuggingFace, Ollama"]

      %% Return path
      E --> D
      F --> D
      H --> D

      D -->|Synthesized Answer| B
      B -->|Conversational Response| A
Loading

7. References