Why NLP processes human language in ML Python - Challenge Your Understanding

Challenge - 5 Problems

🧠 Conceptual

intermediate

Purpose of NLP in understanding human language

Why do Natural Language Processing (NLP) systems process human language?

A To convert human language into machine code for faster computation

B To store human language data without any analysis

C To replace human language with programming languages

D To enable computers to understand and generate human language for communication

💡 Hint

Think about how computers interact with people using language.

NLP helps computers understand and generate human language so they can communicate naturally with people.

🧠 Conceptual

intermediate

Why NLP handles ambiguity in language

Human language is often ambiguous. Why does NLP need to process this ambiguity?

A To correctly interpret the meaning based on context

B To ignore unclear parts and focus only on clear words

C To translate ambiguous words into numbers only

D To remove all ambiguous sentences from data

💡 Hint

Think about how humans understand words with multiple meanings.

NLP processes ambiguity to understand the correct meaning depending on the situation or context.

❓ Predict Output

advanced

Output of tokenizing a sentence

What is the output of this Python code using NLTK to tokenize a sentence?

ML Python

import nltk
nltk.download('punkt', quiet=True)
sentence = "Hello world! NLP processes human language."
tokens = nltk.word_tokenize(sentence)
print(tokens)

A ['Hello', 'world', '!', 'NLP', 'processes', 'human', 'language', '.']

B ['Hello world!', 'NLP processes human language.']

C ['Hello', 'world', 'NLP', 'processes', 'human', 'language']

D ['Hello', 'world!', 'NLP', 'processes', 'human', 'language', '.']

💡 Hint

Tokenization splits words and punctuation separately.

The tokenizer splits words and punctuation marks into separate tokens.

❓ Metrics

advanced

Choosing the right metric for NLP classification

Which metric is best to evaluate an NLP model that classifies emails as spam or not spam when false positives are costly?

A Mean Squared Error

B Recall

C Precision

D Accuracy

💡 Hint

False positives mean marking good emails as spam.

Precision measures how many predicted spam emails are actually spam, reducing false positives.

🔧 Debug

expert

Error in training an NLP model with incorrect input shape

What error will this code raise when training a simple neural network on text data without proper input shape?

ML Python

import tensorflow as tf
model = tf.keras.Sequential([
    tf.keras.layers.Dense(10, activation='relu'),
    tf.keras.layers.Dense(2, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy')

# Input data is a list of strings
texts = ['hello world', 'nlp is fun']
labels = [0, 1]

model.fit(texts, labels, epochs=1)

A SyntaxError: Missing parentheses in call to 'print'

B ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type str)

C TypeError: Cannot call fit on a Sequential model

D No error, model trains successfully

💡 Hint

Model expects numbers, but input is text strings.

The model cannot convert raw text strings directly to tensors without preprocessing.