Univariate Explorer

• Numeric: Continuous or discrete numerical values (shows mean, median, std dev, etc.)
• Categorical: Text or discrete categories (shows frequency counts, mode)
• Date/Time: Temporal data (shows range, patterns)
• Boolean: True/false or binary values (shows proportions)

The app automatically detects data types based on column values.

• Count: Total number of observations
• Missing: Number and percentage of null/empty values
• Mean: Average value (numeric only)
• Median: Middle value when sorted (numeric only)
• Mode: Most frequent value
• Std Dev: Standard deviation, measure of spread
• Variance: Squared standard deviation
• Min/Max: Minimum and maximum values
• Q1/Q3: 25th and 75th percentiles
• Skewness: Measure of asymmetry in distribution
• Kurtosis: Measure of tail heaviness

The histogram shows the frequency distribution of values:

• Bins: Data is grouped into ranges for visualization
• Frequency: Height of bars shows count in each bin
• Shape: Normal, skewed, bimodal, uniform, etc.

Adjust bin count to see more or less detail in the distribution.

Three methods for identifying outliers:

• IQR Method: Values below Q1 - 1.5×IQR or above Q3 + 1.5×IQR
• Z-Score: Values with |z-score| > 3 (>3 standard deviations from mean)
• Modified Z-Score: More robust using median absolute deviation

The box plot visualizes the distribution and highlights outliers.

• EDA: Exploratory data analysis before modeling
• Data Quality: Check for missing values, outliers, and anomalies
• Feature Engineering: Understand variable distributions for transformation
• Reporting: Generate statistical summaries for stakeholders
• Data Profiling: Quick overview of individual variables