Named Entity Recognition basics in ML Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

What is the main goal of Named Entity Recognition (NER)?

Choose the best description of what NER does in text processing.

A Identify and classify key information like names, places, and dates in text.

B Translate text from one language to another automatically.

C Summarize long documents into short paragraphs.

D Detect the sentiment or emotion expressed in a sentence.

💡 Hint

Think about what entities like people or locations are in a sentence.

NER finds and labels important words or phrases such as names of people, places, or organizations in text.

❓ Predict Output

intermediate

Output of a simple NER tagging example

What is the output of this NER tagging code snippet?

ML Python

sentence = "Apple is looking at buying U.K. startup for $1 billion"
entities = [('Apple', 'ORG'), ('U.K.', 'GPE'), ('$1 billion', 'MONEY')]

A [('Apple', 'ORG'), ('U.K.', 'PERSON'), ('$1 billion', 'MONEY')]

B [('Apple', 'PERSON'), ('U.K.', 'ORG'), ('$1 billion', 'DATE')]

C [('Apple', 'ORG'), ('U.K.', 'GPE'), ('$1 billion', 'MONEY')]

D [('Apple', 'ORG'), ('U.K.', 'GPE'), ('$1 billion', 'QUANTITY')]

💡 Hint

ORG means organization, GPE means geopolitical entity, MONEY means monetary value.

The entities are correctly labeled as Apple (organization), U.K. (location), and $1 billion (money).

❓ Model Choice

advanced

Choosing the best model type for NER

Which model type is most suitable for Named Entity Recognition tasks?

A Recurrent Neural Network (RNN) or Transformer-based model trained on sequence labeling

B Convolutional Neural Network (CNN) designed for image classification

C Generative Adversarial Network (GAN) for image generation

D K-Means clustering for unsupervised grouping

💡 Hint

NER requires understanding word order and context in sentences.

RNNs and Transformers handle sequences well and are commonly used for labeling each word in a sentence with entity tags.

❓ Metrics

advanced

Evaluating NER model performance

Which metric is most appropriate to evaluate a Named Entity Recognition model?

A Silhouette score measuring cluster separation

B Accuracy of predicting the exact entity tags for each token

C BLEU score comparing generated text to reference text

D Mean Squared Error between predicted and true entity counts

💡 Hint

NER is a classification task at the token level.

Accuracy or F1-score on token-level entity tags is used to measure how well the model predicts correct entity labels.

🔧 Debug

expert

Debugging incorrect NER predictions

Given this NER model output on the sentence "Barack Obama was born in Hawaii.": Predicted entities: [('Barack', 'PERSON'), ('Obama', 'PERSON'), ('Hawaii', 'ORG')] What is the likely cause of the error?

A The model incorrectly labeled 'Hawaii' as an organization instead of a location.

B The model did not detect any entities in the sentence.

C The model confused the entity type for 'Barack' and 'Obama'.

D The model failed to recognize 'Barack Obama' as a single entity and split it into two.

💡 Hint

Consider how multi-word entities should be tagged in NER.

The model split the full name 'Barack Obama' into two separate entities instead of one combined entity, which is a common error in NER.