Format

CSV files must have a header row as the first line. Column names become attribute names. The last column is treated as the class attribute. No special declarations are required — the parser infers types from the values.

Example

Behavior

Type inference: A column is treated as numeric if every non-empty value parses as a finite number; otherwise it is treated as nominal.

Missing values: Empty cells are counted during preprocessing and receive a maximum-distance penalty during classification and clustering.

Class attribute: The last column is always used as the class. If you need a different column to be the class, reorder your CSV before loading.