Table 3 Effect of data cleaning on the dataset
From: K-means clustering of electricity consumers using time-domain features from smart meter data
Number of rows before processing | 21,000,000 rows |
Consumption (KWH) rows with missing values | 676 rows |
After dropping rows with missing values | 20,999,324 rows |
Outliers detected | 485,003 rows |
After removing the outliers | 20,514,321 |