Finance Product Recommendation System

An AI-powered Client Product Recommendation System designed to predict whether a client will subscribe to a financial product (e.g., a term deposit) based on their profile and historical data. This system leverages feature embeddings, vector search, and ML inference pipelines to provide personalized product recommendations.

---

Features

• Client Profile Analysis: Processes client data including age, balance, campaign interaction, previous product history, and more.
• Hybrid Recommendation: Combines classic ML preprocessing (numeric + categorical features) with embedding-based similarity search for improved recommendations.
• Personalized Prediction: Predicts subscription likelihood with a clear Yes / No verdict.
• Evaluation & Metrics: Outputs evaluation metrics (accuracy, precision, recall, F1-score) and confusion matrices to monitor model performance.
• Extensible Pipeline: Modular preprocessing, inference, and upsert logic for easy experimentation and scaling.
• API Ready: Flask API for serving recommendations in real-time.
• Dockerized: Ready for containerized deployment.

---

Architecture

• Preprocessing Pipeline: Handles numeric scaling, categorical encoding, and binary feature mapping.
• Embedding Generation: Converts client profiles into vector embeddings for similarity search and ML inference.
• Vector Search Database: Uses a vector database (e.g., Qdrant) to store historical client embeddings for fast similarity queries.
• Inference Pipeline: Generates predictions by combining ML models with nearest-neighbor matching in the vector database.
• Flask API / Frontend: Provides endpoints for real-time subscription predictions and batch evaluation.

---

Getting Started

Prerequisites

• Python 3.11+
• Docker (optional, for containerized deployment)
• Vector database setup (e.g., Qdrant)

Installation

1. Clone the repository:

   git clone https://github.com/SilasPenda/Finance_Product_Recommendation_System
   cd Finance_Product_Recommendation_System
   

2. Create & activate virtual environment:

   python -m venv .venv
   source .venv/bin/activate (Linux & Mac)
   ./.venv/Scripts/activate (Windows)
   

3. Install requirements:

   python -m pip install --upgrade pip
   pip install -r requirements.txt
   

4. Create .env and config.yaml files

5. Start App

   python deployment/api.py