Jaccard Similarity | Machine Learning

Author: Arko-Sengupta

machine_learning/jaccard_similarity.py

@@ -0,0 +1,63 @@
+import logging
+
+"""
+Jaccard Similarity Algorithm - Natural Language Processing (NLP) Algorithm
+
+Use Case:
+    - Useful in text analysis and natural language processing (NLP) tasks.
+    - Can be used for document similarity, plagiarism detection,
+and information retrieval.
+
+Dependencies:
+    - Python Standard Library's logging module for error handling.
+"""
+class JaccardSimilarity:
+    def __init__(self) -> None:
+        """
+        Initialize the JaccardSimilarity class.
+        """
+
+    def jaccard_similarity(self, str1: str, str2: str) -> float:
+        """
+        Calculate the Jaccard Similarity between two strings.
+
+        Parameters:
+        - str1 (str): The first string for comparison.
+        - str2 (str): The second string for comparison.
+
+        Returns:
+        - float: The Jaccard similarity between the two strings as a percentage.
+
+        Examples:
+        >>> js = JaccardSimilarity()
+        >>> js.jaccard_similarity("hello world", "hello there")
+        50.0
+        >>> js.jaccard_similarity("apple orange banana", "banana orange apple")
+        100.0
+        >>> js.jaccard_similarity("apple", "banana")
+        0.0
+        """
+        if not str1 or not str2:
+            raise ValueError("Input strings must not be empty.")
+
+        try:
+            set1 = set(str1.split())
+            set2 = set(str2.split())
+
+            intersection = set1.intersection(set2)
+            union = set1.union(set2)
+
+            similarity = len(intersection) / len(union)
+
+            return similarity * 100
+        except Exception as e:
+            logging.error("An Error Occurred: ", exc_info=e)
+            raise e
+
+if __name__ == "__main__":
+    """
+    Main function to Test the Jaccard Similarity between two Texts.
+    """
+    import doctest
+
+    doctest.testmod()  # Run the Doctests