Low missingness
Usually a note to inspect. Confirm whether blanks are expected and document the decision.
Missing values in CSV files
Missing values are blank or unavailable cells that can change how a report, cleanup step, or model should be interpreted.
MLdeck Data Quality reviews missingness in the browser and turns it into readable signals so users can decide what needs review before the file is used downstream.
Why missing values matter
A blank cell can mean many things: a data entry error, an unavailable answer, a field that does not apply, or a value intentionally withheld. Those meanings are different. Treating every blank as the same can lead to weak reports, confusing summaries, or misleading model inputs.
Missingness also affects trust. A column with a few blanks may only need documentation. A column with heavy missingness may need cleanup, exclusion, or a different analysis plan.
How MLdeck reviews missingness
The Data Quality module profiles the CSV locally in the browser. It counts missing cells, summarizes affected columns, and includes missingness in the readiness score and issue list when the signal is strong enough to review.
The result is not an automatic fix. It is a review prompt: which columns have missing values, how much is missing, and whether the missingness may affect the next step.
How to interpret the signal
Moderate missingness
Often needs cleanup discussion. Decide whether to fill, exclude, separate, or explain affected rows.
High missingness
May make a column hard to rely on. Review whether the column should drive reporting or AutoML.
Before reporting, sharing, cleanup, or AutoML
Use missing-value findings as a checklist item. Ask whether blanks have a business meaning, whether they are concentrated in certain columns, and whether they should be documented before the file is shared or modeled.