# Challenge 01 Spaghetti -> Schema ( 2026 ) — Data

来源：https://www.kaggle.com/competitions/challenge-01-spaghetti-schema-2026/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description

## 🧩 Dataset Description

This dataset provides raw operational data from a fictional grocery retailer, designed for the **Challenge 01 – Spaghetti → Schema** competition.
Your task is to normalize these “spaghetti-style” flat files into a clean relational schema suitable for analysis, dashboards, and machine learning workflows.

### 📁 Files Included

| File                                          | Description                                                                                                       |
| --------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| **Products_2024.csv**                         | Product master data including item IDs, names, categories, and supplier information.                              |
| **Promotions_2024.csv**                       | Marketing and discount campaign data including promotion IDs, start/end dates, and participating products.        |
| **Inventory_Snapshots_2024.csv**              | Daily or weekly inventory levels by product and store.                                                            |
| **Sales_2024_Q1.csv** – **Sales_2024_Q4.csv** | Transactional sales data by quarter for 2024. Each record includes product, store, date, units sold, and revenue. |
| **README.txt**                                | Documentation describing column formats, data relationships, and intended normalization guidance.                 |

### 🧱 Objective

Participants must:

1. Analyze the flat files for redundancy and inconsistency.
2. Design a **normalized schema** (e.g., in SQL or ERD form).
3. Provide transformation logic or scripts that produce structured tables (e.g., dimension and fact tables).

This prepares the data for downstream challenges focused on **data integrity, forecasting, and strategy modeling**.

### 🧾 Data Notes

* Data is **synthetic and educational**; it mirrors real-world retail structures but contains no sensitive or private information.
* Field names and patterns are consistent across quarters for joinability.
* Time is represented in local store timestamps but can be aggregated at daily or weekly levels.

### ⚖️ License

**CC BY-NC-SA 4.0** — You may share and adapt the dataset for non-commercial educational or research purposes with attribution.