Page 114 - Touhpad Ai

3. Deep learning is not used in applications like self-driving cars.
4. Computer vision allows a machine to see and interpret images.

5. Voice recognition and speech recognition are the same.

6. NLP applications include email spam filtering and document summarisation.
7. Chatbots use NLP to interact with users and provide answers.

8. AI-driven personalized treatment plans are based on a patient's medical history and lifestyle.
9. AI-powered disaster prediction systems cannot forecast floods or cyclones.

10. AI-powered devices in smartphones cannot recognise faces for security purposes.

SECTION B (Subjective Type Questions)

A. Short answer type questions.
1. Define deep learning.

Ans. Deep learning is a subset of Machine Learning that utilizes neural networks with many layers.
2. List two applications of a Deep Neural Network (DNN).
Ans. Some applications of deep learning are as follows:
Automated driving: Deep learning is used to spot stoplights and traffic signals and also to detect
pedestrians, reducing the incidence of accidents.
Aerospace and defence: Identifying objects from satellites and locate safe and unsafe zones for troops
is another area where deep learning is playing a major role.
3. What do you understand by tokenization?
Ans. Tokenization is the process of breaking text down into smaller, manageable units called tokens, which can
be words, punctuation marks, or subword components.

4. Name the process in NLP that identifies significant entities like people or places in text.
Ans. Named Entity Recognition (NER) identifies key entities like people, locations, or organizations in text.
5. What is the role of AI in accelerating medical research?
Ans. AI accelerates medical research by analysing large datasets to identify patterns, discover new treatments,
and aid in drug development.
B. Long answer type questions.

1. Explain computer vision. Also, explain how to represent images with numbers.
Ans. Computer Vision is a specialised domain dedicated to enabling computers to interpret and analyse digital
images and videos. By employing sophisticated algorithms, it extracts meaningful patterns and information
from visual data—essentially mimicking the human ability to see and understand the world.
Each pixel in a digital image has a numerical value. For black and white images, the value ranges from 0 to
255, where 0 is black and 255 is white. For coloured images, each pixel’s value is based on the RGB colour
model, which stands for Red, Green, and Blue. Each colour channel (Red, Green, Blue) can have a value
from 0 to 255, creating over 16 million possible colours. By mixing different amounts of red, green, and blue,
a wide range of colours can be represented in an image.
2. List the features of speech recognition.
Ans. Some important features include:
∑ ∑ Language weighting: Improves accuracy by giving more importance to commonly used words, such as
product names or technical terms.

112 Touchpad Artificial Intelligence - XI