Normalization (Data Science) 편집하기 (부분)

==Common Methods of Normalization==
Several techniques are commonly used for normalization, each suited to different types of data and model requirements:
*'''Min-Max Scaling''': Scales features to a specified range, typically between 0 and 1.
**Formula: x_scaled = (x - min) / (max - min)
**Suitable for: Data where the minimum and maximum values are known, useful in models sensitive to feature range.

*'''Z-Score Normalization (Standardization)''': Centers data around the mean with a standard deviation of 1, making it comparable across different datasets.
**Formula: x_scaled = (x - mean) / standard deviation
**Suitable for: Data with a normal distribution, commonly used in machine learning models that assume normally distributed features.

*'''Max Abs Scaling''': Scales features by dividing by the maximum absolute value, preserving sign but limiting the range to [-1, 1].
**Formula: x_scaled = x / max(|x|)
**Suitable for: Data with both positive and negative values, ensuring that zero-centered data remains balanced.

*'''Robust Scaling''': Uses the median and interquartile range, making it less sensitive to outliers.
**Formula: x_scaled = (x - median) / IQR
**Suitable for: Data with outliers, providing a more stable normalization for datasets with extreme values.