Entity linking concept in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load a pre-trained entity linking model from spaCy.

NLP

import spacy
nlp = spacy.load('[1]')

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

Aen_core_web_md

Ben_core_web_trf

Cen_core_web_sm

Den_core_entity_linker

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing a general language model like 'en_core_web_sm' which does not include entity linking.

Using transformer models that don't have entity linking by default.

The correct model for entity linking in spaCy is en_core_entity_linker, which includes the entity linking pipeline component.

2fill in blank

medium

Complete the code to extract entities and their linked Wikipedia IDs from a processed document.

NLP

for ent in doc.ents:
    print(ent.text, ent.kb_id_[1])

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

A()

B_

C[]

D.

Attempts:

3 left

💡 Hint

Common Mistakes

Trying to call kb_id_ as a method with parentheses.

Using dot or brackets incorrectly after kb_id.

The attribute kb_id_ is accessed with a trailing underscore and no parentheses or brackets because it is a string property of the entity span.

3fill in blank

hard

Fix the error in the code to add the entity linker component to the spaCy pipeline.

NLP

nlp = spacy.load('en_core_web_sm')
nlp.add_pipe('[1]', last=True)

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

Atextcat

Bner

Centity_linker

Dparser

Attempts:

3 left

💡 Hint

Common Mistakes

Adding 'ner' instead of 'entity_linker'.

Using component names unrelated to entity linking.

The entity linker component is named entity_linker and must be added explicitly to the pipeline if not included by default.

4fill in blank

hard

Fill both blanks to create a dictionary of entity texts mapped to their Wikipedia URLs.

NLP

entity_links = {ent.text: 'https://en.wikipedia.org/wiki/' + ent.kb_id_[1] for ent in doc.ents if ent.kb_id_[2] != ''}

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

A_

B.

C()

D[]

Attempts:

3 left

💡 Hint

Common Mistakes

Using parentheses or brackets which cause errors.

Using dot notation incorrectly after kb_id.

The kb_id_ attribute is accessed as a string property with a trailing underscore and no parentheses or brackets.

5fill in blank

hard

Fill all three blanks to filter entities with confidence score above 0.85 and create a dictionary of their texts and Wikipedia URLs.

NLP

entity_links = {ent.text: 'https://en.wikipedia.org/wiki/' + ent.kb_id_[1] for ent in doc.ents if ent.kb_id_[2] != '' and ent._.kb_[3] > 0.85}

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

A_

B.

Cscore

Did

Attempts:

3 left

💡 Hint

Common Mistakes

Using dot instead of underscore for kb_id_.

Using wrong attribute name for confidence score.

Access kb_id_ as a property with underscore, and the confidence score is stored in the custom extension attribute kb_score accessed as ent._.kb_score.