Marketing Guru Chatbot with Llama3, LangChain, and Logistic Regression

This repository contains the Marketing Guru Chatbot, an AI-powered marketing assistant designed to assist Marketing team members in predicting customer churn rates. The application leverages a conversational interface powered by the llama3 chat model via the LangChain framework and use Logistic regression model to predict the churn rate

Table of Contents

• Features
• Architecture
  • Classification Model Pipeline
  • Proof of Concept (POC) Demo Pipeline
  • System Microservice Pipeline

Features

• Advanced NLP: Utilizes Llama3 for natural language understanding and generation.
• Conversational AI: Built with LangChain for seamless and dynamic conversations.
• Predictive Analytics: Employs Logistic Regression for making data-driven marketing predictions.
• User-friendly Interface: Easy-to-use interface for marketing professionals and enthusiasts.
• API for microservice deployment:

Architecture

Classification Model Pipeline

This pipeline is responsible for training the Logistic Regression model used for predictive analytics.

1. Customer Churn Dataset: The input dataset containing customer data and churn information.
2. Exploratory Data Analysis (EDA): Analyze the dataset to understand the structure, patterns, and relationships within the data.
3. Feature Engineering: Transform raw data into meaningful features that improve the performance of the machine learning model.
4. Logistic Regression: Train a Logistic Regression model using the engineered features.
5. Save Model: Save the trained model as Model.pkl for later use.

Proof of Concept (POC) Demo Pipeline

This pipeline demonstrates the chatbot's capabilities and workflow.

1. User Interaction: The user interacts with the chatbot through a web interface built with Streamlit.
2. Streamlit: Handles user queries and responses, making API calls to Llama3.
3. Llama3: Processes the API calls and generates responses based on the chat history.
4. Chat History: Maintains a cache of the conversation history for context.
5. LangChain System Prompt: Utilizes the Logistic Regression model (Model.pkl) to provide coefficients and intercepts that aid in generating system prompts.

System Microservice Pipeline

This pipeline outlines the microservice architecture of the system.

1. User Interaction: The user sends queries and receives responses through a front-end UI.
2. API Gateway: Routes query requests and responses between the front-end UI and the Marketing Guru microservice.
3. Consul Service Registry: Manages the availability and addresses of microservices.
4. Marketing Guru Assistant Microservice: Contains the Docker image with all necessary components (e.g., classification model.pkl, flask app.py, requirement.txt) to run the chatbot service.