Standardized RFM Feature Dataset for Customer Clustering (XYZ Print MSME: May 2024 – October 2025)

This dataset contains standardized Recency, Frequency, and Monetary (RFM) features generated from the transaction history of a digital printing Micro, Small, and Medium Enterprise (MSME), anonymized as XYZ Print, covering transactions from May 2024 to October 2025. The dataset consists of 2,194 unique customer profiles (CustomerID). Each RFM attribute has been preprocessed using scikit-learn's StandardScaler (z-score normalization: mean = 0, variance = 1) to prepare the data directly for distance-based machine learning algorithms such as K-Means, DBSCAN, or Hierarchical Clustering. Dataset Summary Total Instances: 2,194 unique customers Feature Set: 4 variables (CustomerID, Recency, Frequency, Monetary) Scaling Method: StandardScaler Data Dictionary CustomerID: Unique anonymized identifier assigned to each customer (e.g., P0008, P0010). Recency: Standardized score of the number of days since the customer's last purchase relative to the cutoff date (October 31, 2025). Lower values indicate more recent transactions. Frequency: Standardized score representing the total count of distinct purchase transactions made within the 18-month period. Monetary: Standardized score representing the total monetary expenditure (in IDR) generated by the customer. Pipeline & Preprocessing Notes Raw Feature Calculation: Original R, F, and M metrics were aggregated per CustomerID from the raw sales logs. Standardization: Features were transformed using StandardScaler to ensure mean = 0 and standard deviation = 1 across all numerical columns (Recency, Frequency, Monetary), mitigating feature scale bias during model training. Recommended Applications Machine Learning: Distance-sensitive customer segmentation algorithms (K-Means Clustering, Agglomerative Clustering, Gaussian Mixture Models). Outlier Detection: Identification of high-value anomaly accounts (e.g., bulk business clients). Academic Research: Benchmarking MSME customer behavior in digital printing and local manufacturing sectors.