NER with NLTK in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to tokenize the sentence before named entity recognition.

NLP

import nltk
sentence = "Apple is looking at buying U.K. startup for $1 billion"
tokens = nltk.word_tokenize([1])
print(tokens)

Drag options to blanks, or click blank then click option'

Atext

Bnltk

Ctokens

Dsentence

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the tokenizer function itself instead of the sentence string.

Passing the token list instead of the sentence string.

The word_tokenize function needs the sentence string as input to split it into words.

2fill in blank

medium

Complete the code to tag parts of speech for the tokens.

NLP

pos_tags = nltk.pos_tag([1])
print(pos_tags)

Drag options to blanks, or click blank then click option'

Atokens

Bsentence

Cpos_tags

Dnltk

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the original sentence string instead of tokens.

Passing the POS tags variable itself.

The pos_tag function requires a list of tokens to assign part-of-speech tags.

3fill in blank

hard

Fix the error in the code to perform named entity recognition on POS-tagged tokens.

NLP

named_entities = nltk.ne_chunk([1])
print(named_entities)

Drag options to blanks, or click blank then click option'

Asentence

Bpos_tags

Ctokens

Dnamed_entities

Attempts:

3 left

💡 Hint

Common Mistakes

Passing raw tokens instead of POS-tagged tokens.

Passing the original sentence string.

The ne_chunk function requires POS-tagged tokens as input to identify named entities.

4fill in blank

hard

Fill both blanks to extract named entity labels and their word tokens from the tree.

NLP

for subtree in named_entities:
    if hasattr(subtree, '[1]') and subtree.label() == '[2]':
        print('Entity:', ' '.join([token for token, pos in subtree.leaves()]))

Drag options to blanks, or click blank then click option'

Alabel

Blabel()

CPERSON

Dleaves

Attempts:

3 left

💡 Hint

Common Mistakes

Using label() inside hasattr which expects a string attribute name.

Confusing method call with attribute name.

We check if the subtree has the label attribute and compare it to the entity type string, e.g., PERSON. Here, the blank expects the attribute name label.

5fill in blank

hard

Fill all three blanks to create a dictionary of named entities and their types.

NLP

entities = { ' '.join([token for token, pos in subtree.leaves()]): subtree.[1]() for subtree in named_entities if hasattr(subtree, '[2]') }
print(entities)

# Filter only entities of type [3]

Drag options to blanks, or click blank then click option'

Alabel

CPERSON

Dleaves

Attempts:

3 left

💡 Hint

Common Mistakes

Using leaves instead of label for entity type.

Passing method call as string in hasattr.

Using wrong entity type string.

The dictionary comprehension extracts entity names as keys and their labels as values. We check for the label attribute and call label() to get the entity type. Finally, we filter for entities of type PERSON.