Chapter 4ToIt can be automated.

What is automation?
Automation is any process driven automatically that reduces and eventually eliminates the
need for human intervention.

Automation was once limited to the manufacturing industry. Highly repetitive tasks
how automotive assembly derived into machines and thus the modern assembly line was born. The
machines are excellent at repeating the same task without getting tired and without the errors that humans do
prone to committing in these jobs. This offers greater production, due to the machines
they can work 24 hours without interruptions. The machines also provide a product that is more
uniform.

IoT opens a new world where tasks that previously required human intervention can be
automated returns. As we have seen, IoT allows for the collection of large amounts of data that
they can be analyzed quickly to provide information that can help guide an event or
process.

By adopting the benefits of IoT, automation becomes increasingly important. Access to a

a huge amount of data processed quickly by sensors made people start to think
How to apply the concepts of machine learning and automation to everyday tasks.
Many routine tasks are automated to improve their accuracy and efficiency.

Automation is often associated with the field of robotics. Robots are used in conditions
dangerous activities such as mining, fighting fires, and cleaning up industrial accidents, which reduces the
risk for humans. They are also used in tasks such as automated assembly lines.

Now we see automation everywhere, from self-checkout registers in stores and

the environmental controls of automatic buildings to cars and autonomous vehicles. How many
automated systems find in a single day?

Click on Play in the video to see examples of automation.

Clickhereto read a transcript of the video.

When things start to think

Can things think? Can a device learn from its environment? In this context, there are many
Definitions of the word 'think'. One possible definition is the ability to connect a series of parts.
related information and use it to alter a course of action.

For example, when we are little, we do not have the concept that fire is hot and that putting the
Hand over the fire can cause pain. The fire may seem visually pleasing and actually do.
that one tries to touch the flames. We quickly learn that fire can cause injuries. So
we began to connect the image of fire with the pain it produces. From then on we began to
Think about the results of touching fire and base our actions on this acquired information.

Many devices now incorporate smart technology to modify their behavior in

certain circumstances. This can be as simple as when a smart device reduces its
energy consumption during peak demand periods or as complex as driving a car in a way
autonomous.

Every time a decision or course of action is taken using a device based on information
external, this device is then recognized as a smart device. Currently, many devices
with which we interact carry the word intelligent in the name. This indicates that the device has the
ability to alter its behavior according to its environment.

What technology and smart devices have you interacted with today?
What is artificial intelligence and machine learning?
Artificial intelligence (AI) is the intelligence demonstrated by machines. It is different from intelligence
natural which is the intelligence shown by living organisms. AI uses intelligent agents that can
perceive the environment and make decisions that maximize the probability of achieving a goal or objective
specific. AI refers to systems that mimic the cognitive functions normally associated with
the human mind, like learning and problem solving.

Some of the tasks that are currently considered to require a degree of AI are cars.
autonomous, intelligent routing in content distribution networks, strategic games and
military simulations.

As technology develops, many of the tasks that previously required AI have become
in routine. Many of these tasks migrated from AI to machine learning (ML).

ML is a subset of AI that uses statistical techniques to give computers the ability

to 'learn' from their environment. This allows computers to improve their performance on a task
punctual without being specifically scheduled for that task.

It is especially useful when the design and programming of specific algorithms are difficult or
inviable. Examples of such tasks in computer science include code detection.
malicious, intrusion detection in networks, optical character recognition, recognition of
computer voice and computer vision.

One goal of learning is to be able to generalize based on experience. For machines, this implies the
ability to accurately perform new and previously unseen tasks after gaining experience
with a training dataset. The training dataset should come from data that
representative results of the larger dataset. This dataset allows the machine to create a
general model of these data, which would help her make accurate predictions.

ML in IoT
One of the characteristics of IoT is that it allows the collection of extremely large sets of data.
that can "teach" programs to respond in certain situations. Some of the most common uses
Common machine learning technology includes the following:

Voice recognition: currently many companies offer digital assistants that you
allow the use of voice to communicate with a computer system. Apple, Microsoft, Google, and Amazon
they offer this service. These companies not only allow commands to be provided orally
but also voice-to-text software functions.

Product recommendation: systems create a customer profile and recommend products or

services based on previous patterns. Amazon and eBay users receive
recommendations about products. Organizations like LinkedIn, Facebook, and GooglePlus
users you may want to connect with.

Shape recognition: there are programs that allow diagrams and notes made by hand
in a rudimentary way they are converted into more formal diagrams and text. This allows the shapes and
the handwritten lines are converted into more formal text, which can later be searched and
to analyze itself.

Credit card fraud detection: a profile is created based on purchasing patterns of a

client. Any deviation from these patterns triggers an alert and the system automatically takes action.
action. This action ranges from rejecting the transaction to notifying the authorities. Some of the
events that are detected and could indicate a fraudulent transaction include the purchase of products
that are normally not purchased, purchases in different geographical areas, buy quickly
many different products and buying high-value items.

Facial recognition: security cameras are everywhere, from stores, the

streets, the airports to the transportation centers. These cameras continuously analyze the
 crowds of people, and they usually seek illegal dangerous activities, but they also
can be used to identify and track people. The system creates a pattern of facial features
specific patterns and then looks for a match of these facial patterns that triggers some type of
action.

Think about your interactions with online and offline systems in the past week. How many
What ML applications have you interacted with?

How are ML, AI, and IBN linked?

Intent-based networks leverage the power of automation, AI, and ML to control the
function of a network in order to achieve a specific purpose or intention.

Intent-based networks allow the IT team to specify, in simple language, exactly

what you want the network to do, and it makes it possible. The network is capable of translating intention into policies and,
continuation, use automation to implement the necessary appropriate configurations throughout the network.

Intent-based networks use AI and ML to ensure that any service that

implement meets the required level of service. If it does not meet the level of service, the network based on
the intention can generate alerts and provide suggestions for improvement. In some cases, the network-based
In the intent, it can automatically reconfigure the network to meet service levels.

The intention-based network model shown in the figure consists of three key elements:

Assurance: the assurance element is the end-to-end verification of the

behavior of the entire network. Predicts the outcomes of changes, tracks the
compliance with the original intent, and make recommendations or adjustments in cases where there are
a lack of correspondence between intention and outcome. This stage largely depends on the
AI and ML. The systems are part of a closed loop that continuously monitors performance and
network security, and reconfigures the network to ensure compliance.

Translation: the element of translation is the ability to apply business intent to the
network configuration. The intent is what is desired to be achieved, not the way it is achieved. The intent
it is specified in simple language and is used by the system to create policies throughout said system.
For example, one intention could be to segment guest traffic from corporate traffic, or
allow access to remote users.

Activation: the IBN activation element occurs after it has been specified
intention and the policies have been created. This happens when individual devices are
they provision to align with intention-based policies. It can be a way
automated or semi-automated that allows the network team to verify the configuration beforehand
of the implementation of the devices.

An intention-based network generates an agile and responsive network that extends and
easily adapts to meet business requirements. Uses resources efficiently
highly skilled and allows man and machine to work together to optimize the
customer experience. In addition, intent-based networks provide a safer digital experience
through the automation of complicated and time-consuming processes. This makes it
implementing security policies is much easier.

For more information and free training on intent-based networking, visitCisco

Learning Network.

Use cases for intent-based networks

Objective-based networks allow the company to focus on business goals.
Provide an automated system that understands what the organization needs and makes it happen. Click
Play the video to see an overview of intent-based networks.
Cisco Digital Network Architecture (Cisco DNA) is an example of an intent-based network. It is a
open architecture, that can be expanded and is executed by software. It accelerates and simplifies the
operations of business networks, and reduces costs and risks.

The automation and assurance of Cisco DNA are based on a controller of a network defined by
software (SDN), extensive contextual analysis, network virtualization, and unlimited cloud scalability.

For more information on how Cisco Digital Network Architecture is used in different verticals
from the industry, visithere.

Clickhereto read a transcript of the video.

Summary
This chapter started with the analysis of automation. Automation is any process
automatically driven that reduces and, in the long run, eliminates the need for human intervention. IoT
opens a new world in which tasks that previously required human intervention can be
automated returns. Many devices now incorporate smart technology to modify their
behavior in certain circumstances. Some examples of smart technology can be
find in smart homes and buildings, cities, a smart electric distribution network and
smart cars.

Then, the chapter detailed Artificial Intelligence (AI). AI is the intelligence demonstrated by machines.
As technology develops, many of the tasks that previously required AI have become
routine. Many of these tasks migrated from AI to machine learning (ML). ML is a subset of
AI that uses statistical techniques to give computers the ability to 'learn' from their
environment. Some examples of ML in IoT include facial and voice recognition, recommendation of
products and the detection of credit card fraud.

The next topic of this chapter addressed intent-based networking (IBN). The new enterprise network.
it must integrate IoT devices, cloud-based services, and remote offices in a way that is
relevant and respond to the company. The network must protect these new digital initiatives from the landscape of
constantly changing threats. The IBN represents a systematic approach to linking management
of infrastructures with the business intention.

Finally, in this chapter we analyzed how intention-based networks use AI and ML to

to ensure that any service implemented meets the required level of service. A model of
IBN contains three elements, which include assurance, translation, and activation. Cisco Digital Network
Architecture (Cisco DNA) is an example of an intent-based network. It is an open architecture that
it can be expanded and is executed by software.