Market Basket Analysis is a technique used to analyze the products in customers’ shopping carts to determine which products are sold together. This analysis helps businesses identify cross-selling opportunities, optimize product placement, and better understand customer behavior in general. Within the scope of this project, we will perform Market Basket Analysis using a dirty and real-life dataset.
Dataset Description
The dataset we use in this project includes the products in the customers’ shopping carts, the categories of these products, the date and time of the shopping, and the quantity and price information. The dataset is contaminated with incomplete, incorrect and outlier data, which makes our analyses more realistic and challenging.
TransactionID: Unique ID for each transaction
CustomerID: Unique ID for each customer
ProductID: Unique ID for each product
ProductCategory: The category to which the products belong.
PurchaseDate: The date the purchase was made.
PurchaseTime: The time the purchase was made
Quantity: The amount of product received
Price: Unit price of the product
TotalAmount: Total amount of the purchase (Entered manually and contains errors)
Project Steps
Data Loading and First Look: We will load the dataset and take a general look. We will identify missing, incorrect and outlier data.
Data Cleaning: We will make the dataset suitable for analysis by cleaning missing and erroneous data.
Market Basket Analysis: Using association rule mining techniques (Apriori, FP-Growth), we will analyze which products are purchased together.
Visualization and Interpretation of Results: By visualizing the analysis results, we will gain valuable insights for businesses.
To solve, deeply examine and learn dozens of sales analytics projects like this, you can register now for our 4-week, completely live and project-based Sales Analytics training.
Comments