Logo

Outliers

Upload a dataset to begin

Supported formats: .csv,.xlsx,.xls,.xlsm,.xlsb,.tsv

Not signed in: Up to 20kB per day, no sign-up needed.
Sign in for free for unlimited files, no payment required.

About Outliers

Find data points that don't fit the pattern (Z-Score).

This is a pure detection tool: it computes the mean and standard deviation of a numeric column, then calculates a Z-score for every value to measure how many standard deviations it sits from the average.

Values beyond your chosen threshold (commonly ±2 or ±3) get flagged as statistical anomalies, but nothing in your data is changed, capped, or removed.

It's the diagnostic step, use it to inspect and understand which rows are unusual before deciding what to do about them.

Category: AI & Machine Learning

Common Use Cases

  • Spotting fraudulent transactions with abnormal amounts
  • Flagging sensor readings that indicate equipment malfunction
  • Finding data-entry errors like a typo'd extra zero in a price column

Key Features

  • Z-Score Calculation
  • Configurable Standard Deviation Threshold
  • Row-Level Flagging
  • Works per Numeric Column

Run Outliers first to see which rows are flagged and why, then hand the same column to Outlier Management once you've decided whether to cap, remove, or keep those values.

Frequently Asked Questions

How does it decide what's an outlier?

It uses Z-score, flagging values that fall an unusual number of standard deviations from the mean.

Does it remove outliers automatically?

No, this tool identifies them, use Outlier Management if you want to cap, remove, or flag them in your data.

A first-pass diagnostic for analysts and QA teams auditing a dataset before deeper modeling or reporting.

Read the related guide