/ProductSemanticSearchAI

Primary LanguageJupyter NotebookMIT LicenseMIT

ProductSemanticSearchAI

project assignment from MERCOR [https://work.mercor.io/projects.html]

Semantic Search Engine using AI

This project implements a semantic search engine that uses AI to search for product information based on user input. The search engine is implemented using Python and pandas, and uses a sentence_transformers model to match user queries to product descriptions.

Dataset

The dataset used for this project is a Amazon ML challenge 2023 dataset [https://www.kaggle.com/datasets/ashisparida/amazon-ml-challenge-2023] a product catalog, which is stored in a CSV file named. The dataset contains information about over 2 million products, including their title, bullet points, description, product ID, product type ID, and product length. But to run this on our laptops efficiently I have dropped some of the tuples from the dataset,

Installation

To run this project, you'll need to install the following dependencies:

Python 3
sentence_transformers
Pandas
Tqdm

You can install these dependencies using pip, like so:

pip install sentence_transformers model pandas tqdm

Usage

The script will preprocess the user input and use the learning model to match it to product descriptions in the dataset. The script will then return a list of products that match the user query, along with their descriptions.