product-classification

The 1st place solution for SIGIR 2020 E-Commerce Workshop Multimodal Product Classification Challenge

Implementation of ML algorithms for FlipKart Product Category Classification based on the product's description and other features.

Build a fastText product classification model that can predict a normalized category name for a product, given an unstructured textual representation.

Machine Learning - Multiclass Classification

Categorize and classify anything into a taxonomy or categories using an API that utilizes ChatGPT. Use cases: Classify products into a taxonomy. Classify texts/paragraphs based on predefined categories. Resolve complex taxonomy problems.

End-to-end MLOps pipeline for multimodal e-commerce product classification (text + image) — ingestion, training, inference and monitoring.

API MLOps de classification multimodale Rakuten combinant texte et image avec FastAPI, Docker, DVC, PyTorch et tests automatisés.

Classify e-commerce product descriptions into categories (Household, Books, Electronics, Clothing & Accessories) using SVM and Random Forest models with TF-IDF and Word2Vec representations. Includes data preprocessing, hyperparameter tuning, and model evaluation for performance comparison.

Source Code for User Bias Removal in Fine Grained Sentiment Analysis (CODS-COMAD 2018, DAB@CIKM 2017)

Identification of fashion products using deep learning

NCM (Nomenclatura Comum do Mercosul) codes, their descriptions and hierarchy in formats easy to parse.

CentraleSupélec/OpenClassrooms Data Scientist 2024-2025 - Projet 6

[FRANCAIS]Étude comparative de 10 méthodes d'extraction de features pour e-commerce de luxe | [ENGLISH] Comparative study of 10 feature extraction methods (vision + NLP) for luxury e-commerce classification

Deep Learning for product classification with NLP

A clean, modular, and ML-powered pipeline for grouping and classifying electronic product listings (Laptops & TVs) from noisy vendor specifications using SBERT embeddings, threshold tuning, and confidence scoring. Includes advanced insights, bundle detection, and optional bonus challenges.

Classification de produits avec leurs images et leurs descriptions.

RAG assisted LLM tool and vector embeddings creator for AI enrichment pipeline, to assign GS1 standard categories into messy product data. Fully configurable and adaptable to various tools. Version 2 of the product classifier.

Machine learning technical assessment — classifies marketplace listings as new or used using interpretable rules derived from exploratory analysis. Python + XGBoost, with explainable decision logic.