BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

ARTIFICIAL INTELLIGENCE (AI)

The term Artificial Intelligence (AI) was first invented in the early 1950, and the idea of
artificial intelligence was initially begun by the great computer scientist John McCarthy from 1943
to 1956.

In today's world, technology is growing very fast, and we are getting in touch with
different new technologies day by day.

Here, one of the booming technologies of computer science is Artificial Intelligence which is
ready to create a new revolution in the world by making intelligent machines. The Artificial
Intelligence is now all around us. It is currently working with a variety of subfields, ranging from
general to specific, such as self-driving cars, playing chess, proving theorems, playing music,
Painting, etc.

 AI is one of the fascinating and universal fields of Computer science which has a great
scope in future. AI holds a tendency to cause a machine to work as a human.

Artificial Intelligence is composed of two words Artificial and Intelligence, where Artificial defines
"man-made," and intelligence defines "thinking power", hence AI means "a man-made thinking
power."

AI can be defined as "It is a branch of computer science by which we can create

intelligent machines which can behave like a human, think like humans, and able

Microsoft accmsount 1
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

to make decisions."

 Artificial Intelligence exists when a machine can have human based skills such as
learning, reasoning, and solving problems
 With Artificial Intelligence you do not need to preprogram a machine to do some work,
despite that you can create a machine with programmed algorithms which can work with
own intelligence, and that is the awesomeness of AI.

IMPORTANCE OF AI:
o With the help of AI, you can create such software or devices which can solve real-world
problems very easily and with accuracy such as health issues, marketing, traffic issues,
etc.
o With the help of AI, you can create your personal virtual Assistant, such as Cortana,
Google Assistant, Siri, etc.
o With the help of AI, you can build such Robots which can work in an environment where
survival of humans can be at risk.
o AI opens a path for other new technologies, new devices, and new Opportunities.

ADVANTAGES OF ARTIFICIAL INTELLIGENCE:

Following are some main advantages of Artificial Intelligence:

o High Accuracy with less errors: AI machines or systems are prone to less errors and
high accuracy as it takes decisions as per pre-experience or information.
o High-Speed: AI systems can be of very high-speed and fast-decision making, because of
that AI systems can beat a chess champion in the Chess game.
o High reliability: AI machines are highly reliable and can perform the same action
multiple times with high accuracy.
o Useful for risky areas: AI machines can be helpful in situations such as defusing a
bomb, exploring the ocean floor, where to employ a human can be risky.
o Digital Assistant: AI can be very useful to provide digital assistant to the users such as AI
technology is currently used by various E-commerce websites to show the products as per
customer requirement.

Microsoft accmsount 2
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

o Useful as a public utility: AI can be very useful for public utilities such as a self-driving
car which can make our journey safer and hassle-free, facial recognition for security
purpose, Natural language processing to communicate with the human in human-
language, etc

DISADVANTAGES OF ARTIFICIAL INTELLIGENCE:

Every technology has some disadvantages, and the same goes for Artificial intelligence. Being
so advantageous technology still, it has some disadvantages which we need to keep in our mind
while creating an AI system. Following are the disadvantages of AI:

o High Cost: The hardware and software requirement of AI is very costly as it requires
lots of maintenance to meet current world requirements.
o Can't think out of the box: Even we are making smarter machines with AI, but still
they cannot work out of the box, as the robot will only do that work for which they are
trained, or programmed.
o No feelings and emotions: AI machines can be an outstanding performer, but still it
does not have the feeling so it cannot make any kind of emotional attachment with
human, and may sometime be harmful for users if the proper care is not taken.
o Increase dependency on machines: With the increment of technology, people are
getting more dependent on devices and hence they are losing their mental capabilities.
o No Original Creativity: As humans are so creative and can imagine some new ideas
but still AI machines cannot beat this power of human intelligence and cannot be creative
and imaginative

APPLICATIONS OF AI:
• SPEECH RECOGNITION: It is also known as automatic speech recognition
(ASR), computer speech recognition, or speech-to-text, and it is a capability which uses
natural language processing (NLP) to process human speech into a written format. Many
mobile devices incorporate speech recognition into their systems to conduct voice search
—e.g. Siri—or provide more accessibility around texting.
• CUSTOMER SERVICE: Online virtual agents are replacing human agents along the
customer journey. They answer frequently asked questions (FAQs) around topics, like
shipping, or provide personalized advice, cross-selling products or suggesting sizes for
users, changing the way we think about customer engagement across websites and social
media platforms. Examples include messaging bots on e-commerce sites with virtual
agents, messaging apps, such as Slack and Facebook Messenger, and tasks usually done
by virtual assistants and voice assistants.

Microsoft accmsount 3
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

• COMPUTER VISION: This AI technology enables computers and systems to derive

meaningful information from digital images, videos and other visual inputs, and based on
those inputs, it can take action. This ability to provide recommendations distinguishes it
from image recognition tasks. Powered by convolutional neural networks, computer
vision has applications within photo tagging in social media, radiology imaging in
healthcare, and self- driving cars within the automotive industry.
• RECOMMENDATION ENGINES: Using past consumption behavior data, AI
algorithms can help to discover data trends that can be used to develop more effective
cross-selling strategies. This is used to make relevant add-on recommendations to
customers during the checkout process for online retailers.
• AUTOMATED STOCK TRADING: Designed to optimize stock portfolios, AI-
driven high-frequency trading platforms make thousands or even millions of trades per day
without human intervention.

 INTRODUCTION TO AI ON AZURE
INTRODUCTION TO AI:
1. What is AI?
Artificial Intelligence is a way of making a computer, a computer-controlled robot, or a
software think intelligently, in the similar manner the intelligent humans think.
What is Azure?
Azure is a cloud computing platform and an online portal that allows you to access and
manage cloud services and resources provided by Microsoft.

1.1 FIVE KEY WORKLOADS OF AI

a) Machine-Learning: We "teach" a computer-model to make prediction and draw
conclusions from data.
b) Anomaly-Detection: The capability of software to automatically detect errors or
unusual activity in a system.
c) Computer-Vision: The capability of software to interpret the world visually through
cameras, video and images.
d) Natural Language Processing: The capability of software to interpret written or
spoken language, and respond in kind.
e) Knowledge Mining: The capability of software to extract information from large
volumes of unstructured-data. The extracted-information is used to createa searchable-
knowledge-store.

Microsoft accmsount 4
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

2. UNDERSTAND MACHINE-LEARNING (ML)

Machine learning: Machine learning is a branch of Artificial Intelligence and computer
science which focuses on the use of data and algorithms to imitate the way that humans learn,
gradually improving its accuracy.

2.1. How Machine-Learning works?

• The machines learn from data.
• In today's world, huge volumes of data is created.

• Data is also created by millions of sensors in homes, cars, cities and factories.
• Data-scientists can use the data to train ML-models.
• The models can make predictions and inferences based on the
relationships in the data.

For example, suppose an environmental conservation organization wants volunteers to identify

and catalog different species of wildflower using a phone app
The below procedure is followed:

1. A team of botanists and scientists collect data on wildflower samples.

2. The team labels the samples with the correct species.
3. The labeled data is processed using an algorithm that finds relationships between the features
of the samples and the labeled species.
4. The results of the algorithm are encapsulated in a model.
5. When new samples are found by volunteers, the model can identify the correct species
label.
or e.g. Text messages, emails, social media posts, photographs and videos.

Microsoft accmsount 5
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Microsoft accmsount 6
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

2.2 MACHINE-LEARNING IN MICROSOFT-AZURE

• Microsoft-Azure provides the Machine-Learning (ML) service.
• ML-service is a cloud-based platform for creating, managing, and publishing ML-
models.

FEATURE CAPABILITY
Automated machine learning This feature enables non-experts to quickly create an
effective machine learning model from data.

Azure Machine Learning designer A graphical interface enabling no-code

development of machine learning solutions.

Data and compute management Cloud-based data storage and compute resources that
professional data scientists can use to run data experiment
code at scale.

Pipelines Data scientists, software engineers, and IT operations

professionals can define pipelines to orchestrate model
training, deployment, and management tasks.

3). UNDERSTAND ANOMALY-DETECTION (AD)

The capability to automatically detect errors or unusual activity in a system. Anomaly detection is a technique
that uses AI to identify abnormal behavior as compared to an established pattern. Anything that deviates from an
established baseline pattern is considered an anomaly.
• Anomaly-detection analyses data over time and identifies unusual changes.
• For example,

Microsoft accmsount 7
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

How anomaly-detection can be used in the racing car scenario?

The below procedure is followed:
1) In the car, sensors measure telemetry such as engine-revolutions, brake- temperature
etc.
2) The model is trained to understand expected-range in the telemetry- measurements
over time
3) If a measurement occurs outside of the normal expected-range, the model reports an
anomaly. This anomaly can be used to alert the race-engineer to fix the issue.

3.1 ANOMALY-DETECTION IN MICROSOFT-AZURE:

AD service provides an API [Application Programming Interface] that developers can
use to create anomaly-detection solutions.

4. UNDERSTAND COMPUTER-VISION (CV):

Computer Vision (CV) is a field of Artificial Intelligence (AI) that deals with
computational methods to help computers understand and interpret the content of digital images
and videos. Hence, computer vision aims to make computers see and understand visual data input
from cameras or sensors.
Computer Vision is an area of AI that deals with visual processing.
For example,
The Seeing AI app is a great example of the power of computer vision. Designed for the
blind and low vision community, the Seeing AI app harnesses the power of AI to open up the
visual world and describe nearby people, text and objects.

Microsoft accmsount 8
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

4.1 COMPUTER-VISION MODELS AND CAPABILITIES

• Most Computer-Vision solutions are based on ML-models.
• The model is trained with visual-input from cameras, videos, or images.
The common tasks are:

1) IMAGE CLASSIFICATION: Image classification involves training a machine learning

model to classify images based on their contents. For example, in a traffic monitoring solution
you might use an image classification model to classify images based on the type of vehicle they
contain, such as taxis, buses, cyclists, and so on.

1) OBJECT DETECTION: Object detection machine learning models are trained to

classify individual objects within an image, and identify their location with a bounding box. For
example, a traffic monitoring solution might use object detection to identify the location of
different classes of vehicle.

1) SEMANTIC SEGMENTATION: Semantic segmentation is an advanced machine

learning technique in which individual pixels in the image are classified according to the object
to which they belong. For example, a traffic monitoring solution might overlay traffic images with
"mask" layers to highlight different vehicles using specific colors.

Microsoft accmsount 9
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

1) IMAGE ANALYSIS: You can create solutions that combine machine learning models with
advanced image analysis techniques to extract information from images, including "tags" that
could help catalog the image or even descriptive captions that summarize the scene shown in the
image.

2) FACE DETECTION, ANALYSIS, and RECOGNITION: Face detection is a

specialized form of object detection that locates human faces in an image. This can be combined
with classification and facial geometry analysis techniques to recognize individuals based on
their facial features.

Microsoft accmsount 10
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

3) OPTICAL CHARACTER RECOGNITION (OCR): Optical character recognition is

a technique used to detect and read text in images. You can use OCR to read text in photographs
(for example, road signs or store fronts) or to extract information from scanned documents such
as letters, invoices, or forms.

4.2 COMPUTER-VISION SERVICES IN MICROSOFT-AZURE

SERVICE CAPABILITIES
Computer Vision You can use this service to analyze images and video, and
extract descriptions, tags, objects, and text.

Custom Vision Use this service to train custom image classification and object
detection models using your own images.

Face The Face service enables you to build face detection and
facial recognition solutions.
Form Recognizer Use this service to extract information from scanned forms
and invoices.

Microsoft accmsount 11
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

6. UNDERSTAND NATURAL LANGUAGE PROCESSING (NLP):

Natural language processing (NLP) is a branch of artificial intelligence within
computer science that focuses on helping computers to understand the way that humans
write and speak.
NLP deals with creating software that understands written and spoken language.

Four Tasks of NLP software:

i) Analyse and interpret text in documents, email messages and other sources.
ii) Interpret spoken language and create speech responses.
iii) Automatically translate spoken or written phrases between languages.
iv)Interpret commands and determine appropriate actions.

5.1 Natural Language Processing In Microsoft-Azure:

SERVICE CAPABILITIES
Language Use this service to access features for understanding and
analyzing text, training language models that can understand
spoken or text-based commands, and building intelligent
applications.
Translator Use this service to translate text between more than 60
languages.
Speech Use this service to recognize and synthesize speech, and to
translate spoken languages.
Azure Bot This service provides a platform for conversational AI, the
capability of a software "agent" to participate in a
conversation. Developers can use the Bot Framework to
create a bot and manage it with Azure Bot Service -
integrating back-end services like Language, and connecting
to channels for web chat, email, Microsoft Teams, and others.

Microsoft accmsount 12
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

6. UNDERSTAND KNOWLEDGE MINING (KM): Knowledge mining is

the term used to describe solutions that involve extracting information from large volumes of often
unstructured data to create a searchable knowledge store.

6.1 Knowledge Mining In Microsoft-Azure

One of these knowledge mining solutions is Azure Cognitive Search, a private, enterprise,
search solution that has tools for building indexes. The indexes can then be used for internal only
use, or to enable searchable content on public facing internet assets.
Azure Cognitive Search can utilize the built-in AI capabilities of Azure Cognitive Services
such as image processing, content extraction, and natural language processing to perform knowledge
mining of documents. The product's AI capabilities makes it possible to index previously
unsearchable documents and to extract and surface insights from large amounts of data quickly.

7. CHALLENGES AND RISKS WITH AI:

Artificial Intelligence is a powerful tool that can be used to greatly benefit the world.
However, like any tool, it must be used responsibly.

CHALLENGE OR RISK EXAMPLE

Bias can affect results A loan-approval model discriminates by
gender due to bias in the data with which it
was trained
Errors may cause harm An autonomous vehicle experiences a system
failure and causes a collision
Data could be exposed A medical diagnostic bot is trained using
sensitive patient data, which is stored
insecurely
Solutions may not work for everyone A home automation assistant provides no
audio output for visually impaired users

Users must trust a complex system An AI-based financial tool makes investment
recommendations - what are they based on?

Who's liable for AI-driven decisions? An innocent person is convicted of a crime

based on evidence from facial recognition –
who's responsible?

Microsoft accmsount 13
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

7.1. Understand Responsible AI

At Microsoft, AI software development is guided by a set of six principles, designed to ensure
that AI applications provide amazing solutions to difficult problems without any unintended negative
consequences.

 FAIRNESS: The systems should treat all people fairly. For

example,
Suppose you want to create a model for loan approval application. The model should
predict whether the loan should be approved or denied without bias.

 RELIABILITY AND SAFETY:

AI systems should perform reliably and safely. For example, consider an AI-based
software system for an autonomous vehicle; or a machine learning model that diagnoses
patient symptoms and recommends prescriptions. Unreliability in these kinds of systems
can result in substantial risk to human life.
 PRIVACY AND SECURITY:
AI systems should be secure and respect privacy. The machine learning models on
which AI systems are based rely on large volumes of data, which may contain personal
details that must be kept private. Even after the models are trained and the system is in
production, privacy and security need to be considered. As the system uses new data to
make predictions or take action, both the data and decisions made from the data may be
subject to privacy or security concerns.
 INCLUSIVENESS:
AI systems should empower everyone and engage people. AI should bring benefits
to all parts of society, regardless of physical ability, gender, sexual orientation, ethnicity, or
other factors.

 TRANSPARENCY:
AI systems should be understandable. Users should be made fully aware of the
purpose of the system, how it works, and what limitations may be expected.
 ACCOUNTABILITY:
People should be accountable for AI systems. Designers and developers of AI-
based solutions should work within a framework of governance and organizational
principles that ensure the solution meets ethical and legal standards that are clearly
defined.

Microsoft accmsount 14
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 Use visual tools to create machine learning

models with azure machine learning:
1. Introduction

Machine Learning is the foundation for most artificial intelligence solutions. Creating an
intelligent solution often begins with the use of machine learning to train predictive models using
historic data that you have collected.

Azure Machine Learning is a cloud service that you can use to train and manage machine
learning models.

2. What is Machine-Learning?
Machine learning is a branch of Artificial Intelligence and computer science which focuses on
the use of data and algorithms to imitate the way that humans learn, gradually improving its
accuracy.

• ML uses mathematics & statistics to create a model that can predict unknown
values.

a) Mathematically, machine-learning is a function (f): The function (f) operates on one or more
features (x) to calculate a predicted-label (y).
i.e., f(x) = y

b) The function (f) depends on a no. of factors including

i) Type of model
ii) Algorithm used to train the model

2.1 TYPES OF MACHINE-LEARNING

a) There are 2 general approaches to machine-learning.
1) Supervised ML and

Microsoft accmsount 15
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

2) Unsupervised ML
b) In both approaches, you train a model to make predictions.
1) Supervised Machine-Learning: is a machine learning
Supervised learning
approach in which a data scientist acts like a tutor and trains the AI system by feeding basic rules
and labeled datasets. The datasets will include labeled input data and expected output results. In
this machine learning method, the system is explicitly told what to look for in the input data.
• You to start with a dataset with known label-values.

 Two types of tasks are:

i) Regression: Regression is used to predict a continuous value; like a price or a sales

total.
ii) Classification: Classification is used to determine a class-label. For
example,
 Binary class-label can classify whether a patient has diabetes or not
 Multi-class-labels can classify text as positive, negative, or neutral

2) Unsupervised Machine-Learning: is a machine

Unsupervised learning
learning technique in which the data scientist lets the AI system learn by observing. The training
dataset will contain only the input data and no corresponding output data.
When compared to supervised learning, this machine learning method requires massive
amounts of unlabeled data to observe, find patterns, and learn. Unsupervised learning could be a
goal in itself, for example, discovering hidden patterns in datasets or a method for feature
learning.

• You to start with a dataset without known label-values.

• Clustering is one type of unsupervised task.

Microsoft accmsount 16
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

i) Clustering: Clustering used to determine labels by grouping similar

information into label-groups.
• For example, grouping different shapes.

3. WHAT IS AZURE MACHINE-LEARNING STUDIO?

Training and deploying an effective machine learning model involves a lot of work, much
of it time-consuming and resource-intensive. Azure Machine Learning is a cloud-based service
that helps simplify some of the tasks it takes to prepare data, train a model, and deploy a
predictive service.
Azure Machine Learning studio is a web portal for machine learning solutions in Azure.
It includes a wide range of features and capabilities that help data scientists prepare data, train
models, publish predictive services, and monitor their usage. To begin using the web portal, you
need to assign the workspace you created in the Azure portal to Azure Machine Learning studio

Microsoft accmsount 17
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

3.1 Azure Machine-Learning Compute

Azure Machine Learning is a service for training and managing machine learning models,
for which you need compute on which to run the training process.
Compute targets are cloud-based resources on which you can run model training and data
exploration processes.
There are four kinds of compute resource you can create:

 COMPUTE INSTANCES: Development workstations that data-scientists can use to

work with data and models.
 COMPUTE CLUSTERS: Scalable clusters of virtual-machines for on- demand
processing of experiment code.
 INFERENCE CLUSTERS: Deployment targets for predictive-services that use your
trained models.
 ATTACHED COMPUTE: Links to existing Azure compute-resources such as
Virtual-machines.

4. WHAT IS AZURE AUTOMATED MACHINE-LEARNING

(AAML)?
Azure Machine Learning includes an automated machine learning capability that automatically
tries multiple pre-processing techniques and model-training algorithms in parallel. These
automated capabilities use the power of cloud compute to find the best performing supervised
machine learning model for your data.

Automated machine learning allows you to train models without extensive data science or
programming knowledge. For people with a data science and programming background, it
provides a way to save time and resources by automating algorithm selection and hyper-
parameter tuning.

You can create an automated machine learning job in Azure Machine Learning studio.

Microsoft accmsount 18
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

In Azure Machine Learning, operations that you run are called jobs. You can configure multiple
settings for your job before starting an automated machine learning run. The run configuration
provides the information needed to specify your training script, compute target, and Azure ML
environment in your run configuration and run a training job.

Microsoft accmsount 19
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

5. UNDERSTAND THE AUTOML PROCESS

The steps in a machine-learning process is as follows:

1. Prepare data: Identify the features and label in a dataset. Pre-process, or clean and
transform, the data as needed.
2. Train model: Split the data into two groups, a training and a validation set. Train a machine
learning model using the training data set. Test the machine learning model for
performance using the validation data set.
3. Evaluate performance: Compare how close the model's predictions are to the known
labels.
4. Deploy a predictive service: After you train a machine learning model, you can deploy the
model as an application on a server or device so that others can use it.

These are the same steps in the automated machine learning process with Azure Machine
Learning:

Prepare data
Machine learning models must be trained with existing data. Data scientists expend a lot
of effort exploring and pre-processing data, and trying various types of model-training algorithms
to produce accurate models, which is time consuming, and often makes inefficient use of expensive
compute hardware.

In Azure Machine Learning, data for model training and other operations is usually encapsulated
in an object called a dataset. You can create your own dataset in Azure Machine Learning
studio.

Microsoft accmsount 20
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Train model
The automated machine learning capability in Azure Machine Learningsupports supervised
machine learning models - in other words, models for which thetraining data includes known
label values. You can use automated machine learningto train models for:

 Classification (predicting categories or classes)

 Regression (predicting numeric values)
 Time series forecasting (predicting numeric values at a future point in time) In

Automated Machine Learning you can select from several types of tasks:

In Automated Machine Learning, you can select configurations for the primary metric, type of model used for
training, exit criteria, and concurrency limits.

Microsoft accmsount 21
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Importantly, AutoML will split data into a training set and a validation set. You can configure the details in the
settings before you run the job.

Evaluate performance:
After the job has finished you can review the best performing model. In this case, you
used exit criteria to stop the job. Thus the "best" model the job generated might not be the best
possible model, just the best one found within the time allowed for this exercise.

 EXPLORE COMPUTER VISION IN MICROSOFT

AZURE
1. INTRODUCTION
Computer vision is an area of artificial intelligence (AI) in which software systems are
designed to perceive the world visually, though cameras, images, and video.
• Computer-vision enables AI applications to "see" the world and make sense of it.
• Computers are capable of processing images; either
 From a live camera feed or
 From digital photographs or videos.
• Image processing is the key to creating software that can emulate human visual perception.

1.1 APPLICATIONS OF COMPUTER-VISION

i) Content Organization:
• Identify people or objects in photos.

Microsoft accmsount 22
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

• Then, organize the photos based on the identification.

• Commonly, photo recognition is used in social media applications.
ii) Text Extraction:
• Analyse images that contain text.
• Then, extract the text into a structured format.
iii) Spatial Analysis:
• Identify objects in a space.
• Then, map the movement of objects within the space.

1.2 Computer-Vision Service (CVS)

• To an AI application, an image is just an array of pixel values.
• These numeric values can be used as features to train models.
• Then, the models make predictions about the image and its contents.
• CVS gives you access to pre-trained computer-vision capabilities.

2. GET STARTED WITH IMAGE ANALYSIS ON AZURE

• CVS provides pre-built computer-vision capabilities.

• The service can analyse images, and return detailed information about the image.
2.1 Azure Resources for Computer-Vision
• To use the CVS, you need to create a resource for it in your Azure subscription.
• You can use either of the following resource types:

 Computer-vision: A specific resource for the CVS.

 Cognitive Services: A general resource that includes CVS and Cognitive services
such as Text Analytics, Translator Text etc.
The resources provide 2 pieces of information:

Microsoft accmsount 23
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 A key that is used to authenticate client applications.

Microsoft accmsount 24
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 An endpoint that provides the HTTP address at which your resource can be
accessed.

2.2 Analysing Images with CVS:

• You can submit images to the CVS to perform a wide range of analytical tasks.
TASKS OF COMPUTER VISION:

1) Describing an Image
• Computer-vision has the ability to
 Analyse an image
 Evaluate the objects that are detected and
 Generate a human-readable phrases that describes objects in the image.
• Consider the following image of the building in New York.
• The returned phrases are listed below in the order of confidence.

 A black and white photo of a city.

 A black and white photo of a large city.
 A large white building in a city.

Tagging Visual Features: The image descriptions can be used to suggest tags for the image. The tags can be
associated with the image as metadata that summarizes attributes of the image. These tags can be associated
with the image as metadata that summarizes attributes of the image; and can be particularly useful if you want to
index an image along with a set of key terms that might be used to search for images with specific attributes
or contents.

Microsoft accmsount 25
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

2) Detecting Objects:
The object detection capability is similar to tagging, in that the service can identify
common objects; but rather than tagging, or providing tags for the recognized objects only, this
service can also return what is known as bounding box coordinates. Not only will you get the
type of object, but you will also receive a set of coordinates that indicate the top, left, width, and
height of the object detected, which you can use to identify the location of the object in the
image.

3) Detecting Brands:
This feature provides the ability to identify commercial brands. The service has an existing
database of thousands of globally recognized logos from commercial brands of products.

When you call the service and pass it an image, it performs a detection task and determine if any of
the identified objects in the image are recognized brands. The service compares the brands against
its database of popular brands spanning clothing, consumer electronics, and many more categories.
If a known brand is detected, the service returns a response that contains the brand name, a
confidence score (from 0 to 1 indicating how positive the identification is), and a bounding box

Microsoft accmsount 26
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

(coordinates) for where in the image the detected brand was found.

For example, in the following image, a laptop has a Microsoft logo on its lid, which is identified and
located by the Computer Vision service.

4) Detecting Faces:
The Computer Vision service can detect and analyze human faces in an image, including the ability
to determine age and a bounding box rectangle for the location of the face(s).

The following example shows an image of a person with their face detected and approximate
age estimated.

5) Categorizing an Image: Computer Vision can categorize images based on their contents.
The service uses a parent/child hierarchy with a "current" limited set of categories. When
analyzing an image, detected objects are compared to the existing categories to determine the best
way to provide the categorization. As an example, one of the parent categories is people_. This
image of a person on a roof is assigned a category of people_.

Microsoft accmsount 27
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

A slightly different categorization is returned for the following image, which is assigned to the category
people group because there are multiple people in the image:

6) Detecting Domain-Specific Content: When categorizing an image, the Computer

Vision service supports two specialized domain models:

 Celebrities - The service includes a model that has been trained to identify thousands of
well-known celebrities from the worlds of sports, entertainment, and business.
 Landmarks - The service can identify famous landmarks, such as the Taj Mahal and the
Statue of Liberty.
The following example shows an image of Mysore palace.

Microsoft accmsount 28
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

7) Optical Character Recognition: This service can use OCR capabilities to detect
printed and handwritten text in images.

 EXPLORE NATURAL LANGUAGE PROCESSING (NLP)

• ANALYZING TEXT is a process where you evaluate different aspects of a document or

phrase, in order to gain insights into the content of that text. For the most part, humans are able to
read some text and understand the meaning behind it. Even without considering grammar rules
for the language the text is written in, specific insights can be identified in the text.

• As an example, you might read some text and identify some key phrases that indicate the main
talking points of the text. You might also recognize names of people or well-known landmarks such
as the Eiffel Tower. Although difficult at times, you might also be able to get a sense for how the
person was feeling when they wrote the text, also commonly known as sentiment.

Text Analytics Techniques

 Statistical analysis of terms used in the text. For example, removing common "stop
words" (words like "the" or "a", which reveal little semantic information about the text).
 Frequency analysis of the remaining words (counting how often each word appears) can
provide clues about the main subject of the text.
 Sentiment analysis: Encoding words or terms as numeric features that can be used to
train a machine learning model. For example, to classify a text document based on the

Microsoft accmsount 29
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

terms it contains. This technique is often used to perform sentiment analysis, in which a
document is classified as positive or negative.
 Applying stemming or lemmatization algorithms to normalize words before
counting them - for example, so that words like "power", "powered", and "powerful" are
interpreted as being the same word.
 Applying linguistic structure rules to analyze sentences - for example, breaking down
sentences into tree-like structures such as a noun phrase, which itself contains nouns,
verbs, adjectives, and so on
 Creating vectorized models that capture semantic relationships between words by
assigning them to locations in n-dimensional space. This modeling technique might, for
example, assign values to the words "flower" and "plant" that locate them close to one
another, while "skateboard" might be given a value that positions it much further away.

Get started with text analysis

• The Language service is a part of the Azure Cognitive Services offerings that can perform
advanced natural language processing over raw text.

Azure resources for the Language service

To use the Language service in an application, you must provision an appropriate resource in your
Azure subscription. You can choose to provision either of the following types of resource:

• A Language resource - choose this resource type if you only plan to use natural language
processing services, or if you want to manage access and billing for the resource separately from
other services.

Microsoft accmsount 30
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

• A Cognitive Services resource - choose this resource type if you plan to use the Language
service in combination with other cognitive services, and you want to manage access and billing
for these services together.

Different capabilities of language services are:

1) Language detection
• Use the language detection capability of the Language service to identify the language in
which text is written. You can submit multiple documents at a time for analysis. For each
document submitted to it, the service will detect

• The language name (for example "English").

• The ISO 6391 language code (for example, "en").

• A score indicating a level of confidence in the language detection.

For example, consider a scenario where you own and operate a restaurant where customers can
complete surveys and provide feedback on the food, the service, staff, and so on. Suppose you
have received the following reviews from customers.

• Review 1: "A fantastic place for lunch. The soup was delicious."

• Review 2: "Comida maravillosa y gran servicio."

• Review 3: "The croque monsieur avec frites was terrific. Bon appetit!"
Document Language ISO 6391 Score
Name Code
Review 1 English en 1.0
Review 2 Spanish es 1.0
Review 3 English en 0.9

• Notice that the language detected for review 3 is English, despite the text containing a mix of
English and French. The language detection service will focus on the predominant language in
the text. The service uses an algorithm to determine the predominant language, such as length of
phrases or total amount of text for the language compared to other languages in the text. The
predominant language will be the value returned, along with the language code. The confidence
score may be less than 1 as a result of the mixed language text.

1) Ambiguous or mixed language content

Microsoft accmsount 31
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

• There may be text that is ambiguous in nature, or that has mixed language content. These
situations can present a challenge to the service. An ambiguous content example would be a case
where the document contains limited text, or only punctuation. For example, using the service to
analyze the text results in a value of unknown for the language name and the language identifier,
and a score of NaN (which is used to indicate not a number).

2) Sentiment analysis
• The text analytics capabilities in the Language service can evaluate text and return sentiment
scores and labels for each sentence. This capability is useful for detecting positive and negative
sentiment in social media, customer reviews, discussion forums and more.

• Using the pre-built machine learning classification model, the service evaluates the text and
returns a sentiment score in the range of 0 to 1, with values closer to 1 being a positive
sentiment. Scores that are close to the middle of the range (0.5) are considered neutral or
indeterminate, with value closer to 0 indicate being negative statement.

• For example, the following two restaurant reviews could be analyzed for sentiment.

• "We had dinner at this restaurant last night and the first thing I noticed was how courteous the
staff was. We were greeted in a friendly manner and taken to our table right away. The table was
clean, the chairs were comfortable, and the food was amazing.

• Our dining experience at this restaurant was one of the worst I've ever had. The service was
slow, and the food was awful. I'll never eat at this establishment again.“

• The sentiment score for the first review might be around 0.9, indicating a positive sentiment;
while the score for the second review might be closer to 0.1, indicating a negative sentiment.

3) Indeterminate sentiment
• A score of 0.5 might indicate that the sentiment of the text is indeterminate, and could result
from text that does not have sufficient context to discern a sentiment or insufficient phrasing. For
example, a list of words in a sentence that has no structure, could result in an indeterminate
score. Another example where a score may be 0.5 is in the case where the wrong language code
was used. A language code (such as "en" for English, or "fr" for French) is used to inform the
service which language the text is in. If you pass text in French but tell the service the language
code is en for English, the service will return a score of precisely 0.5.

4) Key phrase extraction

• Key phrase extraction is the concept of evaluating the text of a document, or documents, and
then identifying the main talking points of the document(s). Consider the restaurant scenario discussed

Microsoft accmsount 32
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

previously. Depending on the volume of surveys that you have collected, it can take a long time to
read through the reviews. Instead, youcan use the key phrase extraction capabilities of the
Language service to summarizethe main points.

• We had dinner here for a birthday celebration and had a fantastic experience. We were greeted
by a friendly hostess and taken to our table right away. The ambiance was relaxed, the food was
amazing, and service was terrific. If you like great food and attentive service, you should try this
place.

• Key phrase extraction can provide some context to this review by extracting the following
phrases:

• Attentive service

• Great food

• Birthday celebration

• Fantastic experience

• Table

• Friendly hostess

• Dinner

• Ambiance

• Place

5) Entity recognition
You can provide the Language service with unstructured text and it will return a list of entities in
the text that it recognizes.

For example, suppose you use the Language service to detect entities in the following restaurant review
extract:

"I ate at the restaurant in Seattle last week."

Entity Type SubType Wikipedia URL

Seattle Location https://en.wikipedia.org/wiki/Seattle
last week DateTime DateRange

Microsoft accmsount 33
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 EXPLORE CONVERSATIONAL AI
INTRODUCTION: Conversational AI is a type of artificial intelligence (AI) that can simulate
human conversation.

It is made possible by natural language processing (NLP), a field of AI that allows

computers to understand and process human language. NLP is used to analyze the meaning of
text and speech, and to generate responses that are appropriate and relevant to the conversation.

How conversational AI works:

Conversational AI works by using a combination of natural language processing (NLP) and
machine learning (ML).

Conversational AI systems are trained on large amounts of data, such as text and speech. This
data is used to teach the system how to understand and process human language. The system then
uses this knowledge to interact with humans in a natural way. It’s constantly learning from its
interactions and improving its response quality over time.

Benefits of conversational AI:

- Reduce costs, and increase productivity and operational efficiency through automation
For example, conversational AI can automate tasks that are currently performed by humans and
thereby reduce human errors and cut costs.

- Deliver better customer experience, achieve higher customer engagement and satisfaction

For example, conversational AI can provide a more personalized and engaging experience by
remembering customer preferences and helping customers 24/7 when no human agents are
around.

Examples of conversational AI
To give a few application examples of the constantly evolving conversational AI technology:

Chatbots: often used in customer service applications to answer questions and provide support.

Virtual assistants: often voice-activated and can be used on mobile devices and smart speakers.

Microsoft accmsount 34
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Text-to-speech software: used to create audiobooks, or generate spoken directions.

Speech recognition software: used to transcribe lectures, create transcripts of phone calls, or
generate automatic captions for videos.
BOT: A bot is an application with a conversational interface.
Some common features of bots include:
• Users interact with a bot by initiating activities in turns.
• Activities are events, such as a user joining a conversation or sending a message.
• Messages can be text, speech, or visual interface elements (such as cards or buttons).
• A flow of activities can form a dialog, in which state is maintained to manage a multi-turn
conversation.
• Activities are exchanged across channels, such as web chat, email, Microsoft Teams, and
others.
• We except organizations to engage with us individually, and to be able to answer complex
questions at a personal level
• Many organizations publish support information that answer through a web browser or
a dedicated app.

RESPONSIBLE AI GUIDELIES FOR BOTS

1. Be transparent about what the bot can (and can’t) do.
2. Make it clear that the user is communicating with bot.
3. Enable the bot to seamlessly hand-off to a human if necessary
4. Ensure the bot respects cultural norms.
5. Ensure the bot is reliable.
6. Respect user privacy.
7. Handle data securely.
8. Ensure the bot meets accessibility standards.
9. Assume accountability for bots actions.
•

Microsoft accmsount 35
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Microsoft accmsount 36
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

BOT Solutions on Microsoft Azure are supported by the following technologies:

• Azure Bot Service: A cloud service that enables bot delivery through one or more
channels, and integration with other services.

• Bot Framework Service: A component of Azure Bot Service that provides a REST API
for handling bot activities.

• Bot Framework SDK. A set of tools and libraries for end-to-end bot
development enabling bot development in a range of programming languages.

Bot templates:

The easiest way to get started with the Bot Framework SDK is to base your new bot on
one the templates it provides:
• Empty Bot - a basic bot skeleton
• Echo Bot - a simple "hello world" sample in which the bot responds to messages by echoing
the message text back to the user.
• Core Bot - a more comprehensive bot that includes common bot functionality, such as
integration with the Language Understanding service.

Conversational AI can be broken down into five core components.

i. Natural language processing NLP is the ability of a computer to understand human

language and respond in a way that is natural for humans. This involves understanding the
meaning of words and the structure of sentences, as well as being able to handle idiomatic
expressions and slang. NLP is used to train computers to understand language.

ii. Machine learning Machine learning is a field of artificial intelligence that enables computers
to learn from data without being explicitly programmed. Machine learning is used to train
computers to understand language, as well as to recognize patterns in data. It is also used to
create models of how different things work, including the human brain.

iii. Text analysis Text analysis is the process of extracting information from text data. This
involves identifying the different parts of a sentence, such as the subject, verb, and object,
different types of words in a sentence, such as nouns, verbs, and adjectives. Text analysis is
used to understand the meaning of a sentence, as well as the relationships between different
words. It is also used to identify the topic of a text, as well as the sentiment (positive or

Microsoft accmsount 37
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

negative) of the text.

Microsoft accmsount 38
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

iv. Computer vision Computer vision is the ability of a computer to interpret and understand
digital images. This involves identifying the different objects in an image, as well as the location
and orientation of those objects Computer vision is used to identify the contents of an image, as
well as the relationships between different objects in the image. It is also used to interpret the
emotions of people in photos, and to understand the context of a photo.

v. Speech recognition Speech recognition is the ability of a computer to understand

human speech. This involves recognizing the different sounds in a
spoken sentence, as well as the grammar and syntax of the sentence. Speech
recognition is used to convert spoken words into text, and to understand the meaning
of the words.

How to create Conversational AI?

i. Start by understanding the use cases and requirements: The first step in creating
conversational AI understands the organization’s specific needs and use cases. What
are you trying to achieve with your chatbot? What type of conversations do you
want it to be able to have?
ii. Choose the right platform and toolkit: There are a number of different platforms and
toolkits that you can use to create conversational AI. Each platform has its own
strengths and weaknesses, so choose the platform that best suits your needs. Some
popular platforms include Microsoft Bot Framework, Amazon Lex, Google Dialog
flow, and IBM Watson.
iii. Build a prototype: Once the requirements have been defined and chosen a platform,
it’s time to start building the prototype. Building a prototype will help to test the chatbot
and iron out any kinks before deploying it to the users.
iv. Deploy and test your chatbot Once the prototype is finished, it’s time to deploy and
test the chatbot. Make sure to test it with a small group of users first to get feedback
and make any necessary adjustments.
v. Optimize and improve your chatbot: The final step is to continually optimize
and improve the chatbot. This can be done by tweaking the algorithms, adding
new features, and collecting user feedback.

Microsoft accmsount 39
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 TUNE MODEL HYPERPARAMETERS- AZURE

MACHINE LEARNING(READING)
TUNE MODEL HYPERPARAMETERS: The Parameters are the configuration
model, which are internal to the model. Hyper parameters are the explicitly specified parameters
that control the training process. Parameters are essential for making predictions. Hyper
parameters are essential for optimizing the model.

The process of finding the optimal configuration is sometimes called tuning.

The component supports the following method for finding the optimum settings for a model:
integrated train and tune.

How to configure Tune Model Hyper parameters

 Train a model by using a parameter sweep
This section describes how to perform a basic parameter sweep, which trains a model by using
the Tune Model Hyper parameters component.

1. Add the Tune Model Hyper parameters component to your pipeline in the designer.
2. Connect an untrained model to the leftmost input.
3. Add the dataset that you want to use for training, and connect it to the middle input of
Tune Model Hyper parameters.

Optionally, if you have a tagged dataset, you can connect it to the rightmost input port
(Optional validation dataset). This lets you measure accuracy while training and tuning.

4. In the right panel of Tune Model Hyper parameters, choose a value

for Parameter sweeping mode. This option controls how the parameters are selected.

 Entire grid: When you select this option, the component loops over a grid
predefined by the system, to try different combinations and identify the best learner.
This option is useful when you don't know what the best parameter settings might be
and want to try all possible combinations of values.
 Random sweep: When you select this option, the component will randomly select
parameter values over a system-defined range. You must specify the maximum
number of runs that you want the component to execute. This option is useful
when you want to increase model performance by using the metrics of your choice
but still conserve computing resources.

5. For Label column, open the column selector to choose a single label column.

Microsoft accmsount 40
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Choose the number of runs:

1. Maximum number of runs on random sweep: If you choose a random sweep, you can
specify how many times the model should be trained, by using a random
combination of parameter values.
6. For Ranking, choose a single metric to use for ranking the models.
When you run a parameter sweep, the component calculates all applicable metrics for the
model type and returns them in the Sweep results report. The component uses separate
metrics for regression and classification models.

However, the metric that you choose determines how the models are ranked. Only the top
model, as ranked by the chosen metric, is output as a trained model to use for scoring.

7. For Random seed, enter an integer number as a pseudo random number generator state
used for randomly selecting parameter values over a pre-defined range. This parameter is
only effective if Parameter sweeping mode is Random sweep.
8. Submit the pipeline.

Results of hyper parameter tuning

When training is complete:

 To view the sweep results, you could either right-click the component, and then select
Visualize, or right-click left output port of the component to visualize.

The Sweep results includes all parameter sweep and accuracy metrics that apply to the
model type, and the metric that you selected for ranking determines which model is
considered "best."

 To save a snapshot of the trained model, select the Outputs+logs tab in the right panel
of the Train model component. Select the Register dataset icon to save the model as a
reusable component.

How a parameter sweep works

When you set up a parameter sweep, you define the scope of your search. The search might use a
finite number of parameters selected randomly. Or it might be an exhaustive search over a
parameter space that you define.

 Random sweep: This option trains a model by using a set number of iterations.

You specify a range of values to iterate over, and the component uses a randomly chosen

Microsoft accmsount 41
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

subset of those values. Values are chosen with replacement, meaning that numbers
previously chosen at random are not removed from the pool of available numbers. So the
chance of any value being selected stays the same across all passes.

 Entire grid: The option to use the entire grid means that every combination is tested.
This option is the most thorough, but it requires the most time.

Iterating over many combinations of settings can be time-consuming, so the component provides
several ways to constrain the process:

 Limit the number of iterations used to test a model.

 Limit the parameter space.
 Limit both the number of iterations and the parameter space.

We recommend that you pipeline with the settings to determine the most efficient method of
training on a particular dataset and model.

Choosing an evaluation metric

At the end of testing, the model presents a report that contains the accuracy for each model so
that you can review the metric results:

 A uniform set of metrics is used for all binary classification models.

 Accuracy is used for all multi-class classification models.
 A different set of metrics is used for regression models.

However, during training, you must choose a single metric to use in ranking the models that
are generated during the tuning process. You might find that the best metric varies, depending
on your business problem and the cost of false positives and false negatives.

Controlling the length and complexity of training

Metrics used for binary classification

 Accuracy is the proportion of true results to total cases.

 Precision is the proportion of true results to positive results.
 Recall is the fraction of all correct results over all results.
 F-score is a measure that balances precision and recall.
 AUC is a value that represents the area under the curve when false positives are plotted on
the x-axis and true positives are plotted on the y-axis.
 Average Log Loss is the difference between two probability distributions: the true one,
and the one in the model.

Microsoft accmsount 42
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Metrics used for regression

 Mean absolute error averages all the errors in the model, where error means the distance
of the predicted value from the true value. It's often abbreviated as MAE.
 Root of mean squared error measures the average of the squares of the errors, and then
takes the root of that value. It's often abbreviated as RMSE.
 Relative absolute error represents the error as a percentage of the true value.
 Relative squared error normalizes the total squared error by dividing by the total
squared error of the predicted values.
 Coefficient of determination is a single number that indicates how well data fits a model. A
value of one means that the model exactly matches the data. A value of zero means that
2 2
the data is random or otherwise can't be fit to the model. It's often called r , R , or r-
squared.

Components that don't support a parameter sweep

Almost all learners in Azure Machine Learning support cross-validation with an integrated
parameter sweep, which lets you choose the parameters to pipeline with. If the learner doesn't
support setting a range of values, you can still use it in cross- validation. In this case, a range of
allowed values is selected for the sweep.

Neural Network Regression component

Neural networks are widely known for use in deep learning and modeling complex
problems such as image recognition, they are easily adapted to regression problems. Any class of
statistical models can be termed a neural network if they use adaptive weights and can
approximate non-linear functions of their inputs. Thus
neural network regression is suited to problems where a more traditional regression model cannot
fit a solution.

Neural network regression is a supervised learning method, and therefore requires

a tagged dataset, which includes a label column. Because a regression model predicts a numerical
value, the label column must be a numerical data type.

You can train the model by providing the model and the tagged dataset as an input to Train
Model. The trained model can then be used to predict values for the new input examples.

Configure Neural Network Regression

Neural networks can be extensively customized. This section describes how to create a model

Microsoft accmsount 43
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

using two methods:

 Create a neural network model using the default architecture

If you accept the default neural network architecture, use the Properties pane to set
parameters that control the behavior of the neural network, such as the number of nodes
in the hidden layer, learning rate, and normalization.

Start here if you are new to neural networks. The component supports many
customizations, as well as model tuning, without deep knowledge of neural networks.

 Define a custom architecture for a neural network

Use this option if you want to add extra hidden layers, or fully customize the network
architecture, its connections, and activation functions.

This option is best if you are already somewhat familiar with neural networks. You use
the Net# language to define the network architecture.

Create a neural network model using the default architecture

1. Add the Neural Network Regression component to your pipeline in the designer. You
can find this component under Machine Learning, Initialize, in the Regression category.

2. option.
 Single Parameter: Choose this option if you already know how you want to
configure the model.
 Parameter Range: Select this option if you are not sure of the best parameters, and
want to run a parameter sweep. Select a range of values to iterate over, and the Tune
Model Hyper parameters iterates over all possible combinations of the settings you
provided to determine the hyper parameters that produce the optimal results.
3. In Hidden layer specification, select fully connected case. This option creates a model using
the default neural network architecture, which for a neural network regression model, has
these attributes:
 The network has exactly one hidden layer.
 The output layer is fully connected to the hidden layer and the hidden layer is
fully connected to the input layer.
 The number of nodes in the hidden layer can be set by the user (default value is
100).

Microsoft accmsount 44
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

4. Indicate how you want the model to be trained, by setting the Create trainer mode
Because the number of nodes in the input layer is determined by the number of
features in the training data, in a regression model there can be only one node in the
output layer.

5. For Number of hidden nodes, type the number of hidden nodes. The default is one
hidden layer with 100 nodes. (This option is not available if you define a custom
architecture using Net#.)
6. For learning rate, type a value that defines the step taken at each iteration, before
correction. A larger value for learning rate can cause the model to converge faster,
but it can overshoot local minima.
7. For Number of learning iterations, specify the maximum number of times the algorithm
processes the training cases.
8. For The momentum, type a value to apply during learning as a weight on nodes from
previous iterations.
9. Select the option, Shuffle examples, to change the order of cases between iterations. If
you deselect this option, cases are processed in exactly the same order each time you
run the pipeline.
10. For Random number seed, you can optionally type a value to use as the seed. Specifying a
seed value is useful when you want to ensure repeatability across runs of the same pipeline.
11. Connect a training dataset and train the model:

 If you set Create trainer mode to Single Parameter, connect a tagged dataset
and the Train Model component.
 If you set Create trainer mode to Parameter Range, connect a tagged dataset and
train the model by using Tune Model Hyper parameters.

12. Submit the pipeline.

Results

After training is complete:

 To save a snapshot of the trained model, select the Outputs tab in the right panel of the
Train model component. Select the Register dataset icon to save the model as a reusable
component.

Microsoft accmsount 45
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

COURSE- 02
DATA ANALYST ASSOCIATE [DA-100]

Introduction

With data and information as the most strategic asset of a business, the underlying challenge that organizations
have today is understanding and using their data to positively affect change within the business. Businesses continue
to struggle to use their data in a meaningful and productive way, which impacts their ability to act.

A retail business should be able to use their vast amounts of data and information in such a way that
impacts the business, including:

 Tracking inventory
 Identifying purchase habits
 Detecting user trends and patterns
 Recommending purchases
 Determining price optimizations
 Identifying and stopping fraud
Data analysis is the process of identifying, cleaning, transforming, and modeling data to discover meaningful and useful
information. The data is then crafted into a story through reports for analysis to support the critical decision-making
process.
As the world becomes more data-driven, storytelling through data analysis is becoming a vital component and aspect
of large and small businesses. It is the reason that organizations continue to hire data analysts.

Data-driven businesses make decisions based on the story that their data tells, and in today's data-driven world, data
is not being used to its full potential, a challenge that most businesses face. Data analysis is, and should be, a critical
aspect of all organizations to help determine the impact to their business, including evaluating customer sentiment,
performing market and product research, and identifying trends or other data insights.

While the process of data analysis focuses on the tasks of cleaning, modeling, and visualizing data, the
concept of data analysis and its importance to business should not be understated. To analyze data, core
components of analytics are divided into the following categories:

 Descriptive

Microsoft accmsount 46
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

 Diagnostic
 Predictive
 Prescriptive
 Cognitive

Descriptive analytics
Descriptive analytics help answer questions about what has happened based on historical data. Descriptive
analytics techniques summarize large datasets to describe outcomes to stakeholders.

By developing key performance indicators (KPIs), these strategies can help track the success or failure
of key objectives. Metrics such as return on investment (ROI) are used in many industries, and
specialized metrics are developed to track performance in specific industries.

An example of descriptive analytics is generating reports to provide a view of an organization's sales and
financial data.

Diagnostic analytics
Diagnostic analytics help answer questions about why events happened. Diagnostic analytics techniques
supplement basic descriptive analytics, and they use the findings from descriptive analytics to discover the
cause of these events. Then, performance indicators are further investigated to discover why these events
improved or became worse. Generally, this process occurs in three steps:

1. Identify anomalies in the data. These anomalies might be unexpected changes in a metric or a
particular market.
2. Collect data that's related to these anomalies.
3. Use statistical techniques to discover relationships and trends that explain these anomalies.

Predictive analytics
Predictive analytics help answer questions about what will happen in the future. Predictive analytics
techniques use historical data to identify trends and determine if they're likely to recur. Predictive
analytical tools provide valuable insight into what might happen in the future. Techniques include a
variety of statistical and machine learning techniques such as neural networks, decision trees, and
regression.

Microsoft accmsount 47
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Prescriptive analytics

Prescriptive analytics help answer questions about which actions should be taken to achieve a goal or target.
By using insights from prescriptive analytics, organizations can make data-driven decisions. This
technique allows businesses to make informed decisions in the face of uncertainty. Prescriptive analytics
techniques rely on machine learning as one of the strategies to find patterns in large datasets. By
analyzing past decisions and events, organizations can estimate the likelihood of different outcomes.

Cognitive analytics
Cognitive analytics attempt to draw inferences from existing data and patterns, derive conclusions based
on existing knowledge bases, and then add these findings back into the knowledge base for future
inferences, a self-learning feedback loop. Cognitive analytics help you learn what might happen if
circumstances change and determine how you might handle these situations.

Inferences aren't structured queries based on a rules database; rather, they're unstructured hypotheses
that are gathered from several sources and expressed with varying degrees of confidence. Effective
cognitive analytics depend on machine learning algorithms, and will use several natural language
processing concepts to make sense of previously untapped data sources, such as call center
conversation logs and product reviews.

these different roles in data and the specific responsibility in the overall spectrum of data discovery and understanding:

 Business analyst
 Data analyst
 Data engineer
 Data scientist
 Database administrator

Business analyst

Microsoft accmsount 48
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

While some similarities exist between a data analyst and business analyst, the key differentiator between
the two roles is what they do with data. A business analyst is closer to the business and is a
specialist in interpreting the data that comes from the visualization. Often, the roles of data analyst and
business analyst could be the responsibility of a single person.

Data analyst
A data analyst enables businesses to maximize the value of their data assets through visualization and
reporting tools such as Microsoft Power BI. Data analysts are responsible for profiling, cleaning, and
transforming data. Their responsibilities also include designing and building scalable and effective data
models, and enabling and implementing the advanced analytics capabilities into reports for analysis. A data
analyst works with the pertinent stakeholders to identify appropriate and necessary data and reporting
requirements, and then they are tasked with turning raw data into relevant and meaningful insights.

A data analyst is also responsible for the management of Power BI assets, including reports,
dashboards, workspaces, and the underlying datasets that are used in the reports. They are tasked with
implementing and configuring proper security procedures, in conjunction with stakeholder
requirements, to ensure the safekeeping of all Power BI assets and their data.
Data analysts work with data engineers to determine and locate appropriate data sources that meet
stakeholder requirements. Additionally, data analysts work with the data engineer and database
administrator to ensure that the analyst has proper access to the needed data sources. The data analyst
also works with the data engineer to identify new processes or improve existing processes for collecting
data for analysis.

Data engineer
Data engineers provision and set up data platform technologies that are on-premises and in the cloud.
They manage and secure the flow of structured and unstructured data from multiple sources. The data
platforms that they use can include relational databases, nonrelational databases, data streams, and file
stores. Data engineers also ensure that data services securely and seamlessly integrate across data
platforms.
Primary responsibilities of data engineers include the use of on-premises and cloud data services and
tools to ingest, egress, and transform data from multiple sources. Data engineers collaborate with
business stakeholders to identify and meet data requirements. They design and implement solutions.

While some alignment might exist in the tasks and responsibilities of a data engineer and a database
administrator, a data engineer's scope of work goes well beyond looking after a database and the server
where it's hosted and likely doesn't include the overall operational data management.

Microsoft accmsount 49
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

A data engineer adds tremendous value to business intelligence and data science projects. When the
data engineer brings data together, often described as data wrangling, projects move faster because
data scientists can focus on their own areas of work.

As a data analyst, you would work closely with a data engineer in making sure that you can access the
variety of structured and unstructured data sources because they will support you in optimizing data
models, which are typically served from a modern data warehouse or data lake.

Both database administrators and business intelligence professionals can transition to a data engineer
role; they need to learn the tools and technology that are used to process large amounts of data.

Data scientist
Data scientists perform advanced analytics to extract value from data. Their work can vary from descriptive
analytics to predictive analytics. Descriptive analytics evaluate data through a process known as exploratory
data analysis (EDA). Predictive analytics are used in machine learning to apply modeling techniques that
can detect anomalies or patterns. These analytics are important parts of forecast models. Descriptive and
predictive analytics are only partial aspects of data scientists' work. Some data scientists might work in the
realm of deep learning, performing iterative experiments to solve a complex data problem by using
customized algorithms.

Anecdotal evidence suggests that most of the work in a data science project is spent on data wrangling
and feature engineering. Data scientists can speed up the experimentation process when data engineers
use their skills to successfully wrangle data.

On the surface, it might seem that a data scientist and data analyst are far apart in the work that they do,
but this conjecture is untrue. A data scientist looks at data to determine the questions that need answers
and will often devise a hypothesis or an experiment and then turn to the data analyst to assist with the
data visualization and reporting.

Database administrator
A database administrator implements and manages the operational aspects of cloud-native and hybrid
data platform solutions that are built on Microsoft Azure data services and Microsoft SQL Server. A
database administrator is responsible for the overall availability and consistent performance and
optimizations of the database solutions. They work with stakeholders to identify and implement the
policies, tools, and processes for data backup and recovery plans.

The role of a database administrator is different from the role of a data engineer. A database
administrator monitors and manages the overall health of a database and the hardware that it

resides on, whereas a data engineer is involved in the process of data wrangling, in other words, ingesting,
transforming, validating, and cleaning data to meet business needs and requirements.

The database administrator is also responsible for managing the overall security of the data, granting and
restricting user access and privileges to the data as determined by business needs and requirements.

Microsoft accmsount 50
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Tasks of a data analyst

A data analyst is one of several critical roles in an organization, who help uncover and make sense of
information to keep the company balanced and operating efficiently. Therefore, it's vital that a data
analyst clearly understands their responsibilities and the tasks that are performed on a near- daily basis.
Data analysts are essential in helping organizations gain valuable insights into the expanse of data that they
have, and they work closely with others in the organization to help reveal valuable information.

The following figure shows the five key areas that you'll engage in during the data analysis process

Prepare
As a data analyst, you'll likely divide most of your time between the prepare and model tasks. Deficient
or incorrect data can have a major impact that results in invalid reports, a loss of trust, and a negative
effect on business decisions, which can lead to loss in revenue, a negative business impact, and more.

Microsoft accmsount 51
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Before a report can be created, data must be prepared. Data preparation is the process of
profiling, cleaning, and transforming your data to get it ready to model and visualize.

Data preparation is the process of taking raw data and turning it into information that is trusted and
understandable. It involves, among other things, ensuring the integrity of the data, correcting wrong or
inaccurate data, identifying missing data, converting data from one structure to another or from one type
to another, or even a task as simple as making data more readable.

Data preparation also involves understanding how you're going to get and connect to the data and the
performance implications of the decisions. When connecting to data, you need to make decisions to
ensure that models and reports meet, and perform to, acknowledged requirements and expectations.

Privacy and security assurances are also important. These assurances can include anonymizing data to avoid
oversharing or preventing people from seeing personally identifiable information when it isn't needed.
Alternatively, helping to ensure privacy and security can involve removing that data completely if it
doesn't fit in with the story that you're trying to shape.

Data preparation can often be a lengthy process. Data analysts follow a series of steps and methods to
prepare data for placement into a proper context and state that eliminate poor data quality and allow it to
be turned into valuable insights.

Model
When the data is in a proper state, it's ready to be modeled. Data modeling is the process of determining
how your tables are related to each other. This process is done by defining and creating relationships
between the tables. From that point, you can enhance the model by defining metrics and adding custom
calculations to enrich your data.

Creating an effective and proper data model is a critical step in helping organizations understand and gain
valuable insights into the data. An effective data model makes reports more accurate, allows the data to
be explored faster and more efficient, decreases time for the report writing process, and simplifies future
report maintenance.

The model is another critical component that has a direct effect on the performance of your report and
overall data analysis. A poorly designed model can have a drastically negative impact on the general
accuracy and performance of your report. Conversely, a well-designed model with well- prepared data
will ensure a properly efficient and trusted report. This notion is more prevalent when you are working
with data at scale.

From a Power BI perspective, if your report is performing slowly, or your refreshes are taking a long
time, you will likely need to revisit the data preparation and modeling tasks to optimize your report.

The process of preparing data and modeling data is an iterative process. Data preparation is the first task
in data analysis. Understanding and preparing your data before you model it will make the modeling step
much easier.

Microsoft accmsount 52
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Visualize
The visualization task is where you get to bring your data to life. The ultimate goal of the visualize task is to
solve business problems. A well-designed report should tell a compelling story about that data, which will
enable business decision makers to quickly gain needed insights. By using appropriate visualizations and
interactions, you can provide an effective report that guides the reader through the content quickly and
efficiently, therefore allowing the reader to follow a narrative into the data.

The reports that are created during the visualization task help businesses and decision makers
understand what that data means so that accurate and vital decisions can be made. Reports drive the
overall actions, decisions, and behaviors of an organization that is trusting and relying on the information
that is discovered in the data.

The business might communicate that they need all data points on a given report to help them make
decisions. As a data analyst, you should take the time to fully understand the problem that the business
is trying to solve. Determine whether all their data points are necessary because too much data can
make detecting key points difficult. Having a small and concise data story can help find insights
quickly.

With the built-in AI capabilities in Power BI, data analysts can build powerful reports, without writing
any code, that enable users to get insights and answers and find actionable objectives. The AI capabilities
in Power BI, such as the built-in AI visuals, enable the discovering of data by asking questions, using the
Quick Insights feature, or creating machine learning models directly within Power BI.

An important aspect of visualizing data is designing and creating reports for accessibility. As you build
reports, it is important to think about people who will be accessing and reading the reports. Reports
should be designed with accessibility in mind from the outset so that no special modifications are
needed in the future.

Many components of your report will help with storytelling. From a color scheme that is complementary
and accessible, to fonts and sizing, to picking the right visuals for what is being displayed, they all come
together to tell that story.

Analyze
The analyze task is the important step of understanding and interpreting the information that is
displayed on the report. In your role as a data analyst, you should understand the analytical

capabilities of Power BI and use those capabilities to find insights, identify patterns and trends,
predict outcomes, and then communicate those insights in a way that everyone can understand.

Advanced analytics enables businesses and organizations to ultimately drive better decisions
throughout the business and create actionable insights and meaningful results. With advanced
analytics, organizations can drill into the data to predict future patterns and trends, identify activities
and behaviors, and enable businesses to ask the appropriate questions about their data.

Microsoft accmsount 53
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Microsoft accmsount 54
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Previously, analyzing data was a difficult and intricate process that was typically performed by data
engineers or data scientists. Today, Power BI makes data analysis accessible, which simplifies the data
analysis process. Users can quickly gain insights into their data by using visuals and metrics directly from
their desktop and then publish those insights to dashboards so that others can find needed information.

This feature is another area where AI integrations within Power BI can take your analysis to the next
level. Integrations with Azure machine learning, cognitive services, and built-in AI visuals will help to
enrich your data and analysis.

Manage
Power BI consists of many components, including reports, dashboards, workspaces, datasets, and more.
As a data analyst, you are responsible for the management of these Power BI assets, overseeing the
sharing and distribution of items, such as reports and dashboards, and ensuring the security of Power BI
assets.

Apps can be a valuable distribution method for your content and allow easier management for large
audiences. This feature also allows you to have custom navigation experiences and link to other assets
within your organization to complement your reports.

The management of your content helps to foster collaboration between teams and individuals. Sharing
and discovery of your content is important for the right people to get the answers that they need. It is also
important to help ensure that items are secure. You want to make sure that the right people have access
and that you are not leaking data past the correct stakeholders.

Proper management can also help reduce data silos within your organization. Data duplication can make
managing and introducing data latency difficult when resources are overused. Power BI helps reduce data
silos with the use of shared datasets, and it allows you to reuse data that you have prepared and modeled.
For key business data, endorsing a dataset as certified can help to ensure trust in that data.

The management of Power BI assets helps reduce the duplication of efforts and helps ensure
security of the data.

Microsoft accmsount 55
 BHARATH MATHA DEGREE COLLEGE | ARTIFICIAL INTELLIGENCE

Microsoft accmsount 56