exploratory-data-analysis

Exploratory data analysis systematically examines tabular datasets to identify structure, distributions, relationships, and anomalies. It ensures data characteristics and quality issues are understood before modeling, preventing downstream errors caused by missing values or outliers.

To profile a CSV dataset, the Skill uses pandas and ydata-profiling to assess structure, data types, and memory usage. It automatically generates a comprehensive HTML report summarizing all findings, including distributions and correlations.

You can detect outliers and missing values by running this Skill on your pandas DataFrame. It identifies and quantifies outliers using statistical methods from scipy, and assesses missing values alongside constant columns and duplicate rows.

Yes, you can visualize correlations between variables. The Skill computes and visualizes correlation matrices using seaborn and matplotlib, alongside quantifying distributions for both numeric and categorical variables during relationship exploration.

Yes, the process handles high cardinality categorical variables. During quality assessment, it identifies high cardinality categoricals alongside missing values, constant columns, and duplicate rows to ensure comprehensive data profiling.

The best way to assess data quality is through systematic profiling that identifies missing values, constant columns, high cardinality categoricals, and duplicate rows. This Skill automates that assessment using pandas to surface quality issues before analysis.