generated from lewagon/ecommerce-template
-
Notifications
You must be signed in to change notification settings - Fork 0
thomaszilliox1/Customer-Segmentation
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"collapsed_sections":["WdZA6TAkgb1_"],"authorship_tag":"ABX9TyOTTKn6dCl2+MF8CGbEw6yo"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":[],"metadata":{"id":"c4A3R-syf6iB"}},{"cell_type":"markdown","source":[""],"metadata":{"id":"5OYyDYkSfVCi"}},{"cell_type":"markdown","source":["# Résumé du projet"],"metadata":{"id":"a1mwCAZuehFD"}},{"cell_type":"markdown","source":["## Contexte"],"metadata":{"id":"5JAEndAmf-kF"}},{"cell_type":"markdown","source":["\n","Sur cette place de marché anglophone, des vendeurs proposent des articles à des acheteurs en postant une photo et une description.\n","il devient nécessaire d'automatiser l'attribution de la catégorie des articles,\n","pour rendre l’expérience utilisateur des vendeurs plus fluide. \n","\n"],"metadata":{"id":"ioZzY7SNgPoZ"}},{"cell_type":"markdown","source":["## Mission"],"metadata":{"id":"CjQ1xzjIgQwh"}},{"cell_type":"markdown","source":["Étudier la faisabilité d'un moteur de classification des articles en différentes catégories, à partir du texte (en anglais) et de l’image.\n"],"metadata":{"id":"hZlzU5esgSOR"}},{"cell_type":"markdown","source":["## Méthodes utilisées"],"metadata":{"id":"WgmwmWlUejet"}},{"cell_type":"markdown","source":["### Classification grâce aux descriptions des articles :"],"metadata":{"id":"WdZA6TAkgb1_"}},{"cell_type":"markdown","source":["+ Méthodes classiques : Bag of Words et TF-IDF\n","+ Méthodes de Word Embedding : \n"," - Word2Vec\n"," - BERT : Tenserflow Hub et Hugging Face Transformers\n"," - USE\n"],"metadata":{"id":"0kXxFtdrgnkO"}},{"cell_type":"markdown","source":["Outils utilisés pour améliorer le score de prédiction : TSNE"],"metadata":{"id":"peb0fcaUhO4r"}},{"cell_type":"markdown","source":["### Classification grâce aux images des articles"],"metadata":{"id":"39CDMWHeel-F"}},{"cell_type":"markdown","source":["- Méthode SIFT : Non concluante\n","- Méthode Conventionnal Neural Network : VGG16, VGG19 et ResNet 50 "],"metadata":{"id":"LRkC9XiDhj1m"}},{"cell_type":"markdown","source":["**Comparaison des résultats :**"],"metadata":{"id":"_Ob-4dKMiCIK"}},{"cell_type":"markdown","source":[""],"metadata":{"id":"9GI1KVG1iJ7I"}},{"cell_type":"markdown","source":["### Mise en place d'une API"],"metadata":{"id":"A5bpWn2KiF_z"}},{"cell_type":"markdown","source":["URL de l'API pour le produit \"champagne\" : https://api.edamam.com/api/food-database/v2/parser?app_id=99ac66db&app_key=1fc39dc71e1d27c4138d43442db37e9d&ingr=champagne&nutrition-type=cooking\""],"metadata":{"id":"TprHDkN2xOQg"}}]}About
This project is focused on segmenting e-commerce customers using unsupervised machine learning models, specifically clustering algorithms.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published