ProSearchAI - AI-Powered Product Search Tool - DEMO 🎬

ProSearchAI is an advanced, semantic search tool that leverages AI to understand user intent and context, providing intelligent product recommendations beyond simple keyword-based matching. Powered by a combination of Google’s Generative AI, vector embeddings, and conversational memory, ProSearchAI makes product discovery intuitive and efficient.

Use Case (Important)

Before ProSearchAI: Sam searches for a monitor with a blue light filter, height-adjustable frame, and VESA mount support, but Amazon's results are filled with irrelevant products, leaving him frustrated.

After ProSearchAI: Sam enters his requirements into ProSearchAI and instantly receives a curated list of monitors that match his exact needs, saving time and effort.

Features

• Semantic Search: Understands the meaning of user queries, retrieving relevant products even with non-exact keyword matches.
• Contextual Query Rephrasing: Rephrases queries based on the conversation history to add contextual relevance.
• Retrieval-Augmented Generation (RAG): Generates responses based on relevant context from previous interactions and retrieved product data.
• Conversational Memory: Tracks conversation history to build on previous queries for a cohesive and personalized search experience.
• Streamlit Interface: Interactive chat-based UI to facilitate intuitive searches and responses.

Tech Stack

• Google Generative AI (Gemini): For generating rephrased queries and contextual responses.
• LangChain: To handle embedding generation, vector storage, and conversation memory.
• Chroma: Persistent vector storage solution for efficient similarity search.
• Streamlit: For creating an interactive chat-based UI.

Requirements

• Python 3.7+
• Required libraries: streamlit, pandas, google-generativeai, langchain, chromadb, dotenv

Installation

1. Clone the repository:

   git clone https://github.com/your-username/ProSearchAI.git
   cd ProSearchAI

2. Install dependencies:

   pip install -r requirements.txt

3. Create a .env file in the project root and add your Google API key:

   GOOGLE_API_KEY=your_google_api_key_here

4. Run the Streamlit application:

   streamlit run app.py

Usage

1. Launch the app by running streamlit run app.py.
2. Input Queries in the chat window. ProSearchAI will process the input, consider the conversation context, and display product recommendations.
3. Review Responses displayed by ProSearchAI, which provide context-aware product suggestions.

How it Works

1. Query Rephrasing: When a user submits a query, ProSearchAI rephrases it based on the conversation history using Google Generative AI.
2. Vector Search: The rephrased query is used to search for semantically similar product descriptions within Chroma's vector store.
3. RAG Response Generation: ProSearchAI formulates a response based on the retrieved products and the user’s specific requirements.
4. Conversational Memory: The conversation history is stored to allow for continuous, context-aware interactions.

About the Team

ProSearchAI is crafted with passion by Team Spambots.