Distance Measures in Data Science with Algorithms

Distance Measures in data science with algorithms

1. Euclidean Distance:

import numpy as np

def euclidean_distance(p1, p2):

    return np.sqrt(np.sum((p1 - p2) ** 2))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Euclidean distance:", euclidean_distance(point1, point2))

#clcoding.com

Euclidean distance: 2.8284271247461903

2. Manhattan Distance:

import numpy as np

def manhattan_distance(p1, p2):

    return np.sum(np.abs(p1 - p2))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Manhattan distance:", manhattan_distance(point1, point2))

#clcoding.com

Manhattan distance: 4

3. Cosine Similarity:

from scipy.spatial import distance

def cosine_similarity(p1, p2):

    return 1 - distance.cosine(p1, p2)

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Cosine similarity:", cosine_similarity(point1, point2))

#clcoding.com

Cosine similarity: 0.9838699100999074

4. Minkowski Distance:

import numpy as np

def minkowski_distance(p1, p2, r):

    return np.power(np.sum(np.power(np.abs(p1 - p2), r)), 1/r)

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Minkowski distance:", minkowski_distance(point1, point2, 3))

#clcoding.com

Minkowski distance: 2.5198420997897464

5. Chebyshev Distance:

import numpy as np

def chebyshev_distance(p1, p2):

    return np.max(np.abs(p1 - p2))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Chebyshev distance:", chebyshev_distance(point1, point2))

#clcoding.com

Chebyshev distance: 2

6. Hamming Distance:

import jellyfish

def hamming_distance(s1, s2):

    return jellyfish.hamming_distance(s1, s2)

# Example usage

string1 = "hello"

string2 = "hallo"

print("Hamming distance:", hamming_distance(string1, string2))

#clcoding.com

Hamming distance: 1

7. Jaccard Similarity:

def jaccard_similarity(s1, s2):

    set1 = set(s1)

    set2 = set(s2)

    intersection = set1.intersection(set2)

    union = set1.union(set2)

    return len(intersection) / len(union)

# Example usage

string1 = "hello"

string2 = "hallo"

print("Jaccard similarity:", jaccard_similarity(string1, string2))

#clcoding.com

Jaccard similarity: 0.6

8. Sørensen-Dice Index:

def sorensen_dice_index(s1, s2):

    set1 = set(s1)

    set2 = set(s2)

    intersection = set1.intersection(set2)

    return (2 * len(intersection)) / (len(set1) + len(set2))

# Example usage

string1 = "hello"

string2 = "hallo"

print("Sørensen-Dice index:", sorensen_dice_index(string1, string2))

#clcoding.com

Sørensen-Dice index: 0.75

9. Haversine Distance:

def haversine_distance(lat1, lon1, lat2, lon2):

    R = 6371.0  # Radius of the earth in km

    dLat = np.deg2rad(lat2 - lat1)

    dLon = np.deg2rad(lon2 - lon1)

    a = np.sin(dLat / 2)**2 + np.cos(np.deg2rad(lat1)) * np.cos(np.deg2rad(lat2)) * np.sin(dLon / 2)**2

    c = 2 * np.arctan2(np.sqrt(a), np.sqrt(1 - a))

    return R * c

# Example usage

print("Haversine distance:", haversine_distance(51.5074, 0.1278, 40.7128, -74.0060))

#clcoding.com

  Input In [14]

    a = np.sin(dLat / 2)**2 + np.cos(np.deg2rad(lat1)) *

                                                         ^

SyntaxError: invalid syntax

10. Mahalanobis Distance:

from scipy.spatial.distance import cdist

def mahalanobis_distance(X, Y):

    return cdist(X.reshape(1,-1), Y.reshape(1,-1), 'mahalanobis', VI=np.cov(X))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Mahalanobis distance:", mahalanobis_distance(point1, point2))

#clcoding.com

Mahalanobis distance: [[1.41421356]]

11. Pearson Correlation:

from scipy.stats import pearsonr

def pearson_correlation(X, Y):

    return pearsonr(X, Y)[0]

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Pearson correlation:", pearson_correlation(point1, point2))

#clcoding.com

Pearson correlation: 1.0

12. Squared Euclidean Distance:

def squared_euclidean_distance(X, Y):

    return euclidean_distance(X, Y)**2

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Squared Euclidean distance:", squared_euclidean_distance(point1, point2))

#clcoding.com

Squared Euclidean distance: 8.000000000000002

13. Jensen-Shannon Divergence:

def jensen_shannon_divergence(X, Y):

    M = 0.5 * (X + Y)

    return np.sqrt(0.5 * (rel_entr(X, M).sum() + rel_entr(Y, M).sum()))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Jensen-Shannon divergence:", jensen_shannon_divergence(point1, point2))

#clcoding.com

Jensen-Shannon divergence: 0.6569041853099059

14. Chi-Square Distance:

def chi_square_distance(X, Y):

    X = X / np.sum(X)

    Y = Y / np.sum(Y)

    return np.sum((X - Y) ** 2 / (X + Y))

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Chi-Square distance:", chi_square_distance(point1, point2))

#clcoding.com

Chi-Square distance: 0.01923076923076923

15. Spearman Correlation:

from scipy.stats import spearmanr

def spearman_correlation(X, Y):

    return spearmanr(X, Y)[0]

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Spearman correlation:", spearman_correlation(point1, point2))

#clcoding.com

Spearman correlation: 0.9999999999999999

16. Canberra Distance:

from scipy.spatial.distance import canberra

def canberra_distance(X, Y):

    return canberra(X, Y)

# Example usage

point1 = np.array([1, 2])

point2 = np.array([3, 4])

print("Canberra distance:", canberra_distance(point1, point2))

#clcoding.com

Canberra distance: 0.8333333333333333