data-cleaner

Handling missing values and outliers in a CSV dataset involves applying imputation strategies like median, forward-fill, or interpolation, and using IQR or Z-score methods to cap or remove outliers. This process standardizes raw inputs and produces a clean dataset for modeling.

The best way to normalize numeric features during preprocessing is to apply standard, min-max, or robust scaling, while fixing data types by converting dates to datetime formats and categoricals. This standardization improves downstream analysis and produces a consistent dataset.

Yes, you can use this data cleaning pipeline on Excel and JSON files. It loads and inspects data from common formats like CSV, Excel, and JSON, showing shape, dtypes, and basic statistics before applying imputation, outlier handling, and normalization.

You automate data cleaning for tabular data by running a pipeline that loads raw files, imputes gaps, handles outliers, standardizes formats, and outputs a cleaned dataset. It automatically produces a processed CSV and a JSON quality report documenting all pipeline decisions.

Data normalization does not happen automatically by default; it is controlled via a parameter. You can explicitly enable standard, min-max, or robust scaling for numeric features during preprocessing to ensure your cleaned dataset meets specific modeling requirements.