https://github.com/tseyongg/dsa3101
https://github.com/tseyongg/dsa3101
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/tseyongg/dsa3101
- Owner: tseyongg
- Created: 2025-03-07T08:56:35.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-03-20T08:06:43.000Z (3 months ago)
- Last Synced: 2025-03-20T09:23:51.638Z (3 months ago)
- Language: Jupyter Notebook
- Size: 40.1 MB
- Stars: 1
- Watchers: 1
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.MD
Awesome Lists containing this project
README
# DSA3101 Group 10: AI-Driven Merchandise Customization Platform for E-commerce
## Description
We have developed an AI-powered platform which enables real-time product customization, allowing realistic visuals to be generated on demand. By ensuring high-fidelity product designs, it enhances the customer experience, improves inventory efficiency, and boosts the competitiveness of our e-commerce platform.
## Set-Up
### Install the required packages from `requirements.txt`
```shell
pip install -r requirements.txt
```
### Clone the repository to your computer
Run this in your terminal:
```shell
cd Desktop
git clone https://github.com/tseyongg/DSA3101.git
```
Then head inside:
```shell
cd DSA3101
```
Or you can just open it manually (whichever you prefer).
### Checking for changes:
To see if there are any updates:
```shell
git status
```
If there are changes, pull the latest version:
```
git pull
```
## About Our Data
### Data Source
Our platform uses the **Amazon Fashion Metadata** and its **5-core** data subset from [**Amazon Review Data (2018)**](https://nijianmo.github.io/amazon/index.html), released by Jianmo Ni at UCSD. These datasets provide key product information and user reviews relevant to fashion items.
The curated data is used to enhance the real-time product customization and visualization experience on our platform, as well as inventory management and pricing optimization.
### Data Dictionary
The data dictionary provides a structured overview of the cleaned dataset, detailing column names, data types, and descriptions.
#### 1. Amazon Fashion (5-core)
The data here is a subset of the data in which all users and items have at least 5 reviews. We use the data here as a reviews table as it contains information on user reviews.
| Column Name | Data Type | Description
|:----------------|:---------:|--------------------------------------------------------------|
| `overall` | Float | Rating of the product |
| `verified` | Boolean | Date of purchase |
| `reviewTime` | Datetime | Time of the review |
| `reviewerID` | Object | ID of the reviewer |
| `asin` | Object | ID of the product |
| `style` | Object | Order status (Pending, Shipped, etc.) |
| `reviewerName` | Object | Name of the reviewer |
| `reviewText` | Object | Text of the review |
| `summary` | Object | Summary of the review |
| `image` | Object | Images that users post after they have received the product |
#### 2. Amazon Fashion (Metadata)
The data here contains product metadata, which includes the following:
| Column Name | Data Type | Description
|:----------------|:---------:|--------------------------------------------------------------|
| `title` | Object | Name of the product |
| `brand` | Object | Brand name |
| `feature` | Object | Bullet-point format features of the product |
| `rank` | Object | Sales rank information |
| `asin` | Object | ID of the product |
| `imageURL` | Object | URL of product image |
| `imageURLHighRes`| Object | URL of high resolution product image |