Jaccard similarity in NLP - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this Jaccard similarity calculation?

Given two sets A = {1, 2, 3, 4} and B = {3, 4, 5, 6}, what is the Jaccard similarity computed by the code below?

NLP

A = {1, 2, 3, 4}
B = {3, 4, 5, 6}
intersection = A.intersection(B)
union = A.union(B)
jaccard_similarity = len(intersection) / len(union)
print(round(jaccard_similarity, 2))

A 0.50

B 0.33

C 0.25

D 0.40

💡 Hint

Recall that Jaccard similarity is the size of the intersection divided by the size of the union of two sets.

The intersection of A and B is {3, 4} with size 2. The union is {1, 2, 3, 4, 5, 6} with size 6. So, similarity = 2/6 = 0.3333, rounded to 0.33. But the code rounds to 2 decimals, so output is 0.33.

🧠 Conceptual

intermediate

Which statement best describes Jaccard similarity?

Choose the best description of what Jaccard similarity measures between two sets.

A The ratio of the size of the intersection to the size of the union of two sets.

B The difference in sizes between two sets.

C The sum of the sizes of two sets.

D The ratio of the size of the union to the size of the intersection of two sets.

💡 Hint

Think about how much two sets overlap compared to their total combined size.

Jaccard similarity is defined as the size of the intersection divided by the size of the union of two sets, measuring their overlap.

❓ Metrics

advanced

What is the Jaccard similarity between these two token sets?

Given two token sets from text documents: doc1 = {'apple', 'banana', 'cherry'} and doc2 = {'banana', 'cherry', 'date', 'fig'}, what is their Jaccard similarity?

A 0.4

B 0.5

C 0.6

D 0.75

💡 Hint

Count the common tokens and total unique tokens.

Intersection is {'banana', 'cherry'} size 2. Union is {'apple', 'banana', 'cherry', 'date', 'fig'} size 5. Similarity = 2/5 = 0.4.

🔧 Debug

advanced

Why does this Jaccard similarity code raise an error?

Consider this code snippet:

def jaccard_similarity(list1, list2):
    intersection = list1 & list2
    union = list1 | list2
    return len(intersection) / len(union)

print(jaccard_similarity(['a', 'b'], ['b', 'c']))

Why does it raise an error?

A The function returns a float but print expects a string.

B The function is missing a return statement.

C Lists do not support the '&' and '|' operators; sets are needed.

D The lists are empty, causing division by zero.

💡 Hint

Check the data types and operators used for intersection and union.

The '&' and '|' operators work only on sets, not lists. Using lists causes a TypeError.

❓ Model Choice

expert

Which model output best matches Jaccard similarity for text similarity?

You have two text documents and want to measure their similarity using Jaccard similarity on token sets. Which model output below correctly computes this similarity?

A

def jaccard(doc1, doc2):
    return len(set(doc1).intersection(set(doc2))) / len(set(doc1).union(set(doc2)))

B

def jaccard(doc1, doc2):
    tokens1, tokens2 = doc1.split(), doc2.split()
    return len(tokens1 & tokens2) / len(tokens1 | tokens2)

C

def jaccard(doc1, doc2):
    set1, set2 = set(doc1), set(doc2)
    return len(set1 & set2) / len(set1 | set2)

D

def jaccard(doc1, doc2):
    set1, set2 = set(doc1.split()), set(doc2.split())
    return len(set1.intersection(set2)) / len(set1.union(set2))

💡 Hint

Consider how to tokenize text properly before computing sets.

Option D splits the documents into words (tokens), converts to sets, then computes intersection and union correctly. Option D treats each character as a token, which is incorrect for word-level similarity. Option D uses list operators '&' and '|' which are invalid. Option D also treats characters as tokens.