Information Technology Management iKingShubh

INFORMATION TECHNOLOGY MANAGEMENT

IN THE ERA OF INDUSTRY 4.0 & INDUSTRY 5.0
Prof. Shubh

COURSE MATERIALS

Class #4
AI & ML in Business

https://bit.ly/3YJM56b
Information Technology Management iKingShubh

Updates of
PROJECT
ASSIGNMENTS
Information Technology Management iKingShubh

Recall
• Industry 1.0 to 5.0

• Need for Information Systems (IS) : Types (TPS, MIS, DSS, ERP),
How IS improves decision-making.

• IT Infrastructure: Components of IT infrastructure (hardware,

software, networking), Role of virtualization (cost-saving and scaling).
• Networking:
• VPN, Routing, IP, LAN (Subnet), CDN

• Blockchain
• Block + Chain, Proof of Work, Applications (DeFi, Smart Contracts)
Information Technology Management iKingShubh

We Will Learn
• AI & ML Definition

• Key AI terminology

• Fundamentals of AI

• Hands-on of creating an AI Application

• Applications in Business
Information Technology Management iKingShubh

INTRODUCTION
Information Technology Management iKingShubh

ARE YOU USING AI IN YOUR DAILY LIFE?

Where?

Other than ChatGPT / Gemini / CoPilot?

Information Technology Management iKingShubh

Can’t figure out?

Really?
Information Technology Management iKingShubh

Let us try to answer it:

By the end of the

Class
Information Technology Management iKingShubh

AI
• AI is the simulation of human • Key Components
intelligence processes by • Learning: Acquiring information

machines, especially computer and rules.

systems. • Reasoning: Using rules to reach

approximate conclusions.
• Originated in the 1950s with
• Self-correction: Improving
pioneers like Alan Turing and
performance over time.
John McCarthy.
Information Technology Management iKingShubh

ML
• ML is a subset of AI that enables systems to learn and improve from
experience without being explicitly programmed.

• Emerged from pattern recognition and the theory that computers can
learn without being programmed to perform specific tasks.

• Enables analysis of massive quantities of data.

Information Technology Management iKingShubh

AI vs ML
• AI is the Broad Concept
• AI encompasses all techniques that enable computers to mimic human intelligence.

• ML is a Subset of AI
• ML focuses on allowing machines to learn from data.

• Practical Implications
• AI Applications:
• Expert systems, game playing.
• ML Applications:
• Predictive analytics, recommendation engines.
Information Technology Management iKingShubh

Key Terminology
• Algorithm: A set of rules or steps used to solve a problem.

• Dataset: A collection of data used for training or testing an AI model.

• Training: The process of teaching an AI model using data.

• Inference: Using the trained model to make predictions on new data.

• Artificial Narrow Intelligence (ANI) : Also known as weak AI. AI systems designed to perform
specific tasks.Do not possess consciousness or general intelligence.

• Generative AI (GenAI) : Category of AI models that generate new content resembling existing
data. Creates text, images, music, or other media. Often uses techniques like generative adversarial
networks (GANs) or transformers.

• Artificial General Intelligence (AGI) : Hypothetical form of AI with broad cognitive abilities.
Possesses the ability to understand, learn, and apply intelligence across various tasks. Matches or
surpasses human cognitive capabilities.
Information Technology Management iKingShubh

Key Terminology
• Machine Learning • Computer Vision

• Structured Data • Supervised Learning

• Unstructured Data • Unsupervised Learning
• Deep Learning / Neural • Training Data
Networks • Test Data
• Natural Language Processing • Model
(NLP)
• Prompt Engineering
• Large Language Model (LLM)
Information Technology Management iKingShubh

Key Terminology
• Machine Learning: • Typically stored in tables with rows
• Subset of artificial intelligence. and columns.

• Algorithms learn from data to make • Easily searchable and analyzable

predictions or decisions. using traditional data tools.

• Does not require explicit • Unstructured Data:

programming for each task. • Data lacking a predefined format or
• Improves performance over time organization.
through experience. • Includes text documents, images,
• Structured Data: audio files, and videos.

• Data organized in a fixed format or • More complex to process and analyze

schema. compared to structured data.
Information Technology Management iKingShubh

Key Terminology
• Deep Learning / Neural Networks: • Involves tasks like language translation,

• Subset of machine learning using artificial sentiment analysis, and speech recognition.

neural networks. • Large Language Model (LLM):

• Consists of multiple layers (deep • AI models trained on vast amounts of text
architectures). data.
• Models complex patterns and representations • Understand and generate human-like
in data. language.
• Effective in processing unstructured data like • Capable of tasks such as translation,
images and speech. summarization, and conversational responses.
• Natural Language Processing (NLP): • Generative Pre-Trained Transformer (GPT) :
• Field of artificial intelligence. • GPT-3
• Enables computers to understand, interpret, • GPT-4
and generate human language.
Information Technology Management iKingShubh

Key Terminology
• Computer Vision: inputs to outputs.
• Area of artificial intelligence focused • Used for tasks like classification and
on visual data. regression.
• Enables computers to interpret and • Unsupervised Learning:
understand images and videos. • Machine learning approach using
• Identifies objects, patterns, and unlabeled data.
actions within visual content. • Models discover hidden patterns or
• Supervised Learning: structures in data.
• Machine learning approach using • Used for clustering and
labeled data. dimensionality reduction.
• Models learn the mapping from
Information Technology Management iKingShubh

Key Terminology
• Training Data: • Model:
• Dataset used to teach machine learning • Mathematical representation or
models. algorithm in machine learning.
• Allows models to learn patterns and • Trained on data to perform specific tasks
relationships within the data. like classification or prediction.

• Test Data: • Prompt Engineering:

• Separate dataset used to evaluate a • Practice of designing and refining input
trained model’s performance. prompts for AI language models.
• Assesses generalization ability on new, • Aims to elicit desired and relevant
unseen data. responses.
• Optimizes the effectiveness of the
interaction with AI systems.
Information Technology Management iKingShubh

AI Agents
• Definition: • Reactive Agents: Respond to

• An AI agent is a system that stimuli. (Auto Vacuum

perceives its environment Cleaner)

and takes actions to • Deliberative Agents: Have a
maximize its chances of model of the world. (Chess
success. Playing)
• Hybrid Agents: Combine both
approaches. (Self Driving cars)
Information Technology Management iKingShubh

AI Hands-on
• Basic sentiment analysis tool using a pre-trained model in Google Colab.

• A practical application of Natural Language Processing (NLP) in business

contexts.

• Instructions:
• Open your web browser and go to Google Colab.

• Sign in with your Google account if prompted.

• Click on "New Notebook" to create a new Colab notebook.

Information Technology Management iKingShubh

AI Hands-on
• Code 1:
!pip install transformers
from transformers import pipeline
# Load pre-trained sentiment analysis model
sentiment_analyzer = pipeline("sentiment-analysis")

• Code part 2:
from IPython.display import display, HTML
from ipywidgets import widgets
def analyze_sentiment(text):
result = sentiment_analyzer(text)[0]
return f"Sentiment: {result['label']}, Confidence: {result['score']:.2f}"
Information Technology Management iKingShubh

AI Hands-on
text_input = widgets.Textarea(
value='',
placeholder='Enter text to analyze',
description='Text:',
disabled=False
)
output = widgets.Output()
button = widgets.Button(description="Analyze Sentiment")
def on_button_clicked(b):
with output:
output.clear_output()
print(analyze_sentiment(text_input.value))
button.on_click(on_button_clicked)
display(text_input, button, output)
Information Technology Management iKingShubh

RECAP
• AI terminologies, Hands-on, Prompt Engineering
• Today’s questions:
• ARE YOU USING AI IN YOUR DAILY LIFE, Other than
ChatGPT / Gemini / CoPilot?

• Alexa, Voice Prompts, Text to voice, Image

search, Friend list, Follow suggestions