Similarity (Data Science) 편집하기 (부분)

==Key Similarity Measures==
Several similarity metrics are commonly used, each suited to different types of data:
*'''Euclidean Distance''': Measures the straight-line distance between two points in Euclidean space. Commonly used for numerical data, as in k-means clustering or k-nearest neighbors (kNN).
  - Formula: d(x, y) = √((x₁ - y₁)² + (x₂ - y₂)² + ... + (xₙ - yₙ)²)
*'''Cosine Similarity''': Calculates the cosine of the angle between two vectors, useful for text data or high-dimensional spaces. Often applied in document similarity and recommendation systems.
  - Formula: cos(x, y) = (Σ(xᵢ * yᵢ)) / (√Σ(xᵢ²) * √Σ(yᵢ²))
*'''Jaccard Similarity''': Measures the overlap between two sets as the ratio of their intersection to their union. Commonly used for binary or categorical data, like measuring similarity between two user profiles.
  - Formula: J(A, B) = |A ∩ B| / |A ∪ B|
*'''Manhattan Distance''': Calculates the sum of the absolute differences between coordinates, also known as "city block" distance. Useful for grid-based data or high-dimensional spaces.
  - Formula: d(x, y) = |x₁ - y₁| + |x₂ - y₂| + ... + |xₙ - yₙ|
*'''Pearson Correlation''': Measures the linear correlation between two variables, commonly used in collaborative filtering in recommendation systems.
  - Formula: r(x, y) = Σ((xᵢ - x̄)(yᵢ - ȳ)) / (√Σ(xᵢ - x̄)² * √Σ(yᵢ - ȳ)²)