CHAPTER 8

Question 1

……………..The ability of systems to learn from data and improve their performance automatically without being explicitly programmed.

The ability to learn:
✓ Something new.
✓ Something new about something you already knew.
✓ How to do something better, either more efficiently or with more
accuracy.

Answer 1

Machine Learning (ML)

Question 2

Types of ML
Question: Is there a teacher?

Answer 2

Supervised Learning – Yes, there is a teacher.: The model is trained on labeled data (input + correct output).
* Learns to predict or classify based on past examples.

Unsupervised Learning – No teacher, learns on its own.: The model explores patterns in unlabeled data.
* Finds hidden structures or groupings.

Semi-Supervised Learning – The teacher helps sometimes.: Mix of labeled and unlabeled data.
* Useful when labeling is expensive or time-consuming

Reinforcement Learning – Learns from experience and feedback.:
* No direct teacher, but gets rewards or penalties based on actions.
* Learns optimal behavior over time through trial and error.

Question 3

T/F NLP is a subset of AI focused on helping machines understand and
work with human language.

Answer 3

T

Question 4

With NLP, machines can :

Answer 4

• Understand spoken or written language.
• Analyze meaning, emotion, or intent.
• Perform speech recognition.
• Generate or summarize text automatically.

Question 5

T/F Machine Learning (ML) for Natural Language Processing (NLP)
and Text Analytics involves using algorithms and narrow AI to help
machines understand the meaning behind text.

⚫ It helps understand and analyze human language by identifying parts
of speech, entities, sentiment, and more. They turn raw, unstructured
text into structured, usable data.
⚫ These technologies enhance accuracy, speed up analysis, and
automate processes—making it possible to analyze massive volumes
of text quickly and effectively.

Answer 5

T

Question 6

Types of Supervised Learning Techniques in NLP : Classification & Sentence Segmentation

Answer 6

T

Question 7

T/F Classification Assigning text to predefined categories based on patterns learned from labeled data.

Answer 7

T

Question 8

Some examples of classification tasks are:

Answer 8

Sentiment Analysis
News Article Classification
Spam Detection

Question 9

T/F During training, a feature extractor is used to convert each input value to a feature set. Pairs of feature sets and labels are fed into the machine learning algorithm to generate a model.

Answer 9

T

Question 10

T/F During prediction, the same feature extractor is used to convert unseen inputs to feature sets. These feature sets are then fed into the model, which generates predicted labels

Answer 10

T

Question 11

Answer 11

DONE

Question 12

………………Dividing long text into individual sentences using labeled data to train
the model.

Answer 12

Sentence Segmentation

Question 13

Some examples Sentence Segmentation tasks are:

Answer 13

• Text Segmentation: Dividing articles, news, or blogs into sentences for
  easier processing.
• Linguistic Analysis: Helps in POS tagging, dependency parsing, and
  machine translation.
• Machine Translation: Accurate sentence division for better translations.

Question 14

…………… involves training a model without pretagging or annotating.

Answer 14

Unsupervised machine learning

Question 15

Some of these techniques are:Unsupervised machine learning

Answer 15

✓ Clustering
✓ Latent Semantic Indexing
✓ Matrix Factorization
✓ Concept Matrix
✓ Syntax Matrix