Synthetic Data Generator

STEP 1

Upload and inspect your dataset

CSV only · Maximum 10 MB

Drop a CSV file here or select it from your device

The file should already be cleaned. Categorical columns will be identified and excluded from modelling.

Missing-value treatment

This treatment will be applied only to the numeric columns used by the model.

Remove incomplete rows Exclude rows containing a missing or infinite value in any selected numeric column. Impute with the median Replace missing or infinite values with the median of the corresponding numeric column.

STEP 2

Configure synthetic generation

Gaussian Copula

Numeric columns

Select at least two non-constant numeric columns.

Multivariate distribution Vine Copula will be enabled in the next development stage. Univariate distribution Synthetic sample size Maximum: three times the usable original rows. Random seed Use the same seed to reproduce the same generation.

Two outputs will be created Raw synthetic values and adjusted values replaced by the closest values observed in each original column.

DATASET SUMMARY

Inspection results

Ready

Original rows 0

Usable rows 0

Numeric columns 0

Excluded columns 0

Maximum synthetic rows 0

Numeric duplicate rows 0

Column	Detected type	Role	Missing	Unique	Minimum	Median	Maximum

CURRENT STAGE

Dataset preparation

This stage detects numeric variables and prepares the rows that can be used by the copula model.

Current rules

Only numeric columns are modelled.
Categorical columns are excluded.
The maximum synthetic sample is three times the usable rows.
Infinite values are treated as missing values.
Files are stored only for temporary processing.

Privacy notice

Synthetic data can reduce direct exposure of original records, but it does not by itself guarantee anonymity or formal privacy.

Synthetic Data Generator

Upload and inspect your dataset

Configure synthetic generation

Synthetic generation results

Adjusted synthetic data

Raw synthetic data

Inspection results

Dataset warnings