AWS Services - Part 1

Amazon AI
An AWS's deeply customizable machine learning and AI platform
Amazon AI is Amazon's Machine Learning and AI platform.
Amazon Rekognition Video is a tool designed to help recognize and track people in real-time video feeds.
Amazon Transcribe will use AI to convert audio recordings into text files.
Amazon Comprehend is able to analyze text and provide information on it such as language, sentiment and
the service can also extract key phrases.
Amazon Translate is a real-time language translation tool that uses machine learning to translate text in a
more natural and understandable way.
AWS also announced its new SageMaker service to aid developers in deploying AI models as well as an
AI-powered camera called DeepLens. Using DeepLens, businesses will be able to develop and test
vision-based AI functions at a faster pace.
Ref Links AI Startups : https://www.cbinsights.com/research/artificial-intelligence-top-startups/
Amazon Machine Learning : contains a series of
visualization tools to analyze data without the need
to learn complex machine learning code.
Amazon Elastic MapReduce (EMR) : A
managed Hadoop framework designed to perform log
analysis, predictive analytics, real-time analytics, and
other big data operations.
Spark : an Apache product that runs inside of Amazon
EMR as a distributed processing system for big data
workloads.

Amazon BlockChain - A Distributed Database for Transaction between 2 Parties.
A Blockchain is a distributed database that allows
the direct transactions betweens two parties
Without the need of centralized authority.

Amazon Internet of Things – The IOT
AWS IOT is a managed cloud platform that lets connected
devices easily and secure internet with cloud application
and other devices.
AWS IOT the applications can keep track of and communicate
with all the devices , all the times and even they are not
connected.
The above Screenshot showing all the devices pertaining to
AWS IOT and depict the purpose of each services.
Ref URL :
https://www.qwiklabs.com/focuses/1804?locale=fr&parent=
catalog#
Use Case : Turn on the Air Conditioning
A Simple IOT device simulator on Amazon EC2 that
will send sensor data(temp) to AWS IOT
Gateway. The Amazon IOT Rule that will publish
a notification to Amazon SNS topic when the
temperature of the device is within the
threshold. By Connecting email address with
Amazon SNS topics , we will receive an email
notification when the threshold met. Finally you
will update the device to “turn on air
conditioning” results in lowering
temperature

Message Queuing and Transport Protocol – MQTT
A Practical Protocol for Internet of Things
MQTT : Designed for minimal network traffic and
constrained devices.
The above tables shows the effective type of Protocol for
IOT.
• MQTT Target large network of small devices that need to
be monitor or controlled from backend server on the
Internet.
• MQTT Server is called a broker and the clients are simply
the connected devices.
• When a device wants to send data to broker, we call this
operation as a “Publish”.
• When a device wants to receive data from broker , we
call this operation as a “Subscribe”.
• Clients are publishing and subscribing topics . So broker
handles the publishing/subscribing actions to the target
topics.
Use Case : Temperature sensor:
IOT Device has a temperature sensors.
The device defines the topics it want to
publish on, ex : Temp then pushes the
message as “temperature values”
The Phone/desktop application subscribe to
the topic “Temp” then it receives the
message device has published.
The Broker role here to take message
“temperature value” and deliver it to
phone/desktop.
MQTT is faster than Http and it’s consume
more power consuming.
Mosquito Broker is an Open Source Message
broker that implements the MQTT
protocol.

Amazon Simple Workflow Service – Use Case Amazon SWF for MER and MSL

Amazon Simple Workflow Service
Amazon Cloud Search is fully managed service make it easy setup, manage and scale a search
solution for your Website or Application.
Key Features :
✔ Run Your Windows Application in the Cloud

Amazon Cloud Search
Amazon Cloud Search is fully managed service
make it easy setup, manage and scale a
search solution for your Website or
Application.
Key Features :
✔ Instant on Browser Access to the APPS as
they need.
✔ One instance per end users – no shared
resources.
✔ Works with existing
apps,identify,entitlements and back-end.
✔ Use multiple App at the same time.
✔ Multiple Storage Options, Support HTML5.
✔ High-fidelity visualization deliver to
browser.
✔ ASE-256 encrypted, streaming via AWS ALB.
✔ HTTP access via streaming gateways.
✔ No ports to open firewall.
✔ Managed Streaming Solutions for desktop.
✔ Move desktop apps to cloud with
no-rewrite.
✔Pay per hour for running in your fleet.
✔Scaling polices and instance type choice optimize cost.
✔Pay Per unique user than connect in a month
✔User fee waived for BYOL (Buy Your Own License RDS
calls).
✔Upload files, test a workflow, save your workflow.
✔Life cycle hooks for streaming instances, a build-in
storage for users.

Amazon APPSTREAM 2.0
Amazon APPSTREAM 2.0 is fully managed
streaming service, centrally managed your
desktop application.
Key Features :
✔ Instant on Browser Access to the APPS as
they need.
✔ One instance per end users – no shared
resources.
✔ Works with existing
apps,identify,entitlements and back-end.
✔ Use multiple App at the same time.
✔ Multiple Storage Options, Support HTML5.
✔ High-fidelity visualization deliver to
browser.
✔ ASE-256 encrypted, streaming via AWS ALB.
✔ HTTP access via streaming gateways.
✔ No ports to open firewall.
✔ Managed Streaming Solutions for desktop.
✔ Move desktop apps to cloud with
no-rewrite.
✔ Pre and post process visualization on AWS.
✔Pay per hour for running in your fleet.
✔Scaling polices and instance type choice optimize cost.
✔Pay Per unique user than connect in a month
✔User fee waived for BYOL (Buy Your Own License RDS
calls).
✔Upload files, test a workflow, save your workflow.
✔Life cycle hooks for streaming instances, a build-in
storage for users.

Amazon Docker vs. CloudFront
• Docker : An Open Platform for Developers and
Sysadmin to Build , Ship and Run Distributed
Application.
• A Container is an running instance of an image
-OS,Software,Application are its components.
• Docker file build with the image runs on
container.
• EC2 : Docker is apart of EC2 Container Service.
• CloudFront : A Global Content Delivery
Network.
Docker
Cloud Front

Amazon Elastic File Storage
The File Storage in AWS Cloud
✔ Amazon EFS provides file storage in the
AWS Cloud. With the help EFS a file can
be created and mounted on Amazon EC2
instance.
✔ Amazon EC2 instance can read and write
data to and from our file system .
✔ Amazon EFS file system also can be
mounted in your Virtual Private Network
through Network File System version 4.0
and 4.1(NFSv4) protocol.
✔ To Access Amazon EFS file in VPC
multiple mount target can be created.
With help of Domain Naming System the
files can be mounted which resolves IP
Address of EFS mount target in the AZ in
an AWS Region.
Benefits of AWS Direct Connection :
AWS Customer’s can transfer large amount
of data or which require high performance
should consider AWS Direct Connection.
Reduce Your Bandwidth Cost
Consistent Network Performance.
Compatible with all AWS Service.
Private Connectivity to your Amazon VPC.

Amazon API Gateway
• Amazon API Gateway is fully managed
service which makes it easy for developer to
create, monitor,maintain,publish and
secure API at any scale.
• Creation of API acts as front door for
application to access data, business logic or
functionality from backend services.
• Key Features :
• API Gateway handles all task involved in
accepting and processing up to hindered of
thousands concurrent API’s including traffic
management,authorization,access control,
Monitoring and API version Management.
• It Allows you simultaneously run multiple
version of same API.
• Facilitates server less application using API
Gateway Socket API and Lambda to send
and receive message to connected users
and devices avoiding the need to
implement complex polling mechanism.
• AWS X-Ray, AWS Cloud Trail and Amazon
CloudWatch are tools used to log, monitor
API execution and management operation.

Amazon CloudFront
A Global Content delivery Network – Deliver Data,Video’s,Application and API
✔ Amazon CloudFront A Global Content
Delivery Network.
✔ Accelerate All Content Type
✔ Dynamic Content such as Web
Application,API’s
✔ Static Content such as image,
video etc.
✔ AWS CloudFront works with AWS
S3 out of box.
Benefits of AWS CloudFront :
Security to the Content.
Integrating Network.
Programmable CDN. (Lambda Edge used for)
Integration with Key AWS Services.
Great Performance and Economical.

The Clouds Comparisons - AWS,AZIRE and Google

Amazon Snowball
A PETABYTE DATA TRANSPORT TO AWS
AWS Snowball is a petabyte scale data
transport . Data is encrypted by snowball
client before it is written to snowball
application.
Key are managed by AWS and never written
to Snowball and it is an 256-bit encryption.
Use Case :
Collect and analyze oceanic and costal
images
60 TB of data per week
Snowball lets OSU migrate TB of data in days
at a fraction of cost.

Amazon Direct Connect
The Dedicated, High Performance, Secure Link to AWS
Amazon Direct Connect establishes a
dedicated network connection from your
premises to AWS.
AWS Direct Connect also establish private
connectivity between AWS and datacenter,
office or co-location environment , using
MLPS or dedicated leased lines avoiding the
internet route.
Benefits of AWS Direct Connection :
AWS Customer’s can transfer large amount
of data or which require high performance
should consider AWS Direct Connection.
Reduce Your Bandwidth Cost
Consistent Network Performance.
Compatible with all AWS Service.
Private Connectivity to your Amazon VPC.
Elastic and Simple
Df

Route53
The Highly Available and Scalable Domain Naming Service by AWS
Amazon Route 53 is a highly available and scalable
DNS service offered by AWS. Like any DNS service,
Route 53 handles domain registration and routes
users’ Internet requests to your application –
whether it’s hosted on AWS or elsewhere.

Amazon EC2 (Elastic Compute Cloud) Resize computer instance in the Cloud
• Amazon EC2 is a web service can be used to
launch and manage instance in Amazon
Datacenter.
• Consuming Minimum time to install a
Software with EC2 than install with traditional
hardware.
• AMI : Amazon Machine Image , a
pre-configured bundle software location and it
is a part of S3 the storage part of Amazon.
• Provides complete control of your computing
resource and let you run on Amazon proven
Computing Environment.
Getting Started with EC2 :
Step 1: Sign up for Amazon EC2
Step 2: Create a key pair
Step 3: Launch an Amazon EC2 instance
Step 4: Connect to the instance
Step 5: Customize the instance
Step 6: Terminate instance and delete the volume
created
Purpose of EC2 :
• Scale as requirement.
• Run application securely.
• Improve resilience and reduce cost.
• Log as root (Linux) /Admin (window).
• Able to create AMI , Start/Stop via
Console or API.
• Instance and Reserved Instance
utilization report for EC2 Usage
Report.

Amazon ECS (Elastic Container Service)
A Logical Group of EC2 Machines/Instances
• ECS a highly Scalable, fast , Container Management
Service that makes it easy to run, stop and manage
Docker as container on a cluster.
• ECS allows you to simplify your views of EC2
instances as pool of resources
• It handles installing Containers, Scaling
• Monitoring and Managing Instances through both
API and AWS Management Console.
Task is the blueprint
Describing which docker
container to run
and represent your
Application.
Service defines max,min
tasks from one task
run at any given
time,autoscaling
and load balancing.

AWS Redshift With TABLEAU
Bring Your On License

Amazon AWS With TABLEAU
https://www.tableau.com/Amazon-Web-Services/Marketplace/Support
• Tableau Software and AWS offers Tableau
Server on Hourly Basis facilitates a solution
that scale up or down with user’s needs.
• Tableau Server for AWS is browser and
mobile-based visual analytics anyone can
use. Publish interactive dashboards with
Tableau Desktop and share them
throughout your organization.
• With Tableau Server in the AWS
Marketplace, you can empower your
organization with live interactive
dashboards without needing to purchase or
administer your own servers. (Pay for
tableau and AWS As Bundle).
• Bring-Your-Own-License (BYOL) option to
launch a virtual EC2 server with Tableau
Server pre-installed.
• Active Directory and other third party
authentication methods will not work with
Marketplace AMIs.
Pay for tableau and AWS
As Bundle
Bring Your Own License

Amazon EC2 Use Case Resize computer instance in the Cloud
Amazon EC2 :
The Amazon EC2 service comes under the compute
domain and it provides services that help to compute
workloads. Amazon EC2 web interface is used to
reduce the expensive physical servers by creating
virtual machines. Also, they help in managing
different features of the virtual servers such as
security, ports, and storage. Amazon EC2 is highly
preferable while creating a virtual server within a few
minutes with just a few clicks.

Amazon Kinesis Video Streams
Amazon Kinesis Video Streams : makes it easy to
securely stream videos from connected to AWS
Analytics, Machine Learning and other processing.
Amazon Kinesis Video Streams is integrated with
Amazon Rekognition Video, making it easy for you to
build applications that take advantage of computer
vision and video analytics. You can also build custom
applications using popular open-source ML
frameworks to process and analyze your video
streams.
Secure : Identity and Access Management the IAM
protect the streaming data with automatic encryption
and it transit the data using TSL protocol.
S3: Amazon kinesis video stream use S3 for storage.
Real-Time and Batch Process : To build real-time
application that uses live stream of data. The stream
of data which is stored in S3 used for batch processing.
Use Cases:
• Smart Home : Facilitate stream video and audio from
camera equipped home device.
Pet monitor as an example.
• Smart City : Cities equipped with traffic camera ,
parking lots helps to solve traffic problem etc.
Amber Alert System as example.
• Industrial Automation: Collection of data generated
from Radar and LIDAR signals, with use of ML
facilitates prediction of the dataset for decision
making.

Simple Notification Service in Amazon
• The architecture includes Amazon SNS to trigger the
processing pipelines when new content is updated,
and Amazon SQS to decouple incoming jobs from
pipeline processor.
• Ref Link :
•
https://aws.amazon.com/partners/success/nasa-im
age-library/
• Pub-Sub Messaging A Message
published to a topic is
immediately received by all of the
subscribers of the topic.
• Amazon SNS : Simple , flexibly
fully managed messaging and
mobile push notification service
for high throughput highly
reliable message delivery.
• Use Case : The NASA Image and
Video Library provides easy
access to more than 140,000 still
images, audio recordings, and
videos—documenting NASA’s
more than half a century of
achievements in exploring the
vast unknown.
`
Ref Link :
https://aws.amazon.com/pub-sub-messaging/
https://aws.amazon.com/messaging/
https://docs.aws.amazon.com/sns/latest/dg/sns-common-scenarios.htm
l
Pub/Sub Messaging

Amazon Workspace
Amazon Workspace Amazon Workspace is a
managed desktop service which is secure
and reliable. It enables the user to securely
access the applications, documents, and the
resources required to ameliorate the
application form anywhere on the devices
Key Features :
✔ Choose your operating system (Windows
or Amazon Linux) and select from a range of
hardware configurations, software
configurations, and AWS regions. For more
information, see Amazon WorkSpaces
Bundles.
✔
Amazon Workspace vs. Amazon Appstream 2.0 :
While the two AWS services are somewhat
similar, it’s important to remember that Amazon
AppStream 2.0 is focused on hosting individual
applications on AWS, while Amazon WorkSpaces
creates virtual desktops that can be used to
create entire working environments for you and
your team.
The bottom line is that if you’re looking to move
your existing legacy applications to AWS, you’ll
want to look at Amazon AppStream 2.0 in more
detail, and if you’re just looking for a quick and
easy way to deploy Windows virtual desktops for
your users, Amazon WorkSpaces is most likely an
ideal solution.
App Stream

Amazon Step Function
Coordinate the components of distributed Application and Micro service
AWS Step Function launched in 2016 intended to co-ordinate components of distributed application
using workflows. Step Function is to simplify by visualizing different component in an
architecture and to manage , organize Lambdas.
API Gateway to set function is a good option is to trigger by new Lambda. The API Gateway or init,
that triggers a init Lambda that will call step function that contains Hello World Lambda.
Sample Step Function Code:
{ "Comment": "An example of the Amazon States Language using a parallel state to execute two branches
at the same time.", "StartAt": "Parallel", "States": { "Parallel": { "Type": "Parallel",
"ResultPath":"$.output", "Next": "Parallel 2", "Branches": [ { "StartAt": "Parallel Step 1, Process 1",
"States": { "Parallel Step 1, Process 1": { "Type": "Task", "Resource":
"arn:aws:lambda:us-west-2:XXXXXXXXXXXX:function:LambdaA", "End": true } }

AWS Redshift vs. Hadoop
• Hadoop is a framework for distributed
processing of large dataset across clusters of
commodity (Open Source from Apache).
• HDFS is master slave architecture consist of
Name Node and Data Node to process or
operate metadata, real data respectively.
• Database operations are configured by
developers
• Hive is DW build on top of HDFS (Shared
Storage) and Map Reduce(Computation).
• There is preemption in Hadoop.
• Hadoop scales as many as petabyte thus by
scaling hadoop doesnot re-shuffling. For
load balancing use rebalance utility.
• Hadoop accepts every data formats and
data type imaginable.
• Data Integration - More flexible to store on
local drive , relational db, or in cloud(S3
included) and an easy import to hadoop.
• Amazon Redshift is Cloud based web
service by Amazon. (Priced Service).
• Redshift provide a console to create and
manage Amazon redshift cluster
• Automates DB Operation by parsing
execution plans.
• AWS Redshift is a data warehouse build
on top of propriety technology developed
by ParAccel , does not use Map Reduce or
HDFS as like Hadoop.
• There is no Preemption in Redshift.
• Largest Redshift node comes with 16TB of
Storage and max of 100 nodes.
• Redshift is strict with data formats.
• Data Integration – Redshift only load data
from Amazon S3 or Dynamo DB, load data
via single thread by default. S3
Ref Link : https://www.educba.com/hadoop-vs-redshift/
https://www.xplenty.com/blog/hadoop-vs-redshift/

• Real Time Architectures :• Stream Based Architecture : Allows us to react
to customer’s online behavior within seconds
while processing billions of events each day.
• Used : In the area of Streaming, Firehouse and
Stream Analysis.
• Streaming is used for custom processing
per incoming record. Terminology below
to know.
• A Shard - Identify group of records in
the stream.
• 1 MB/Sec of data input
• 2 MB/Sec of data output.
• A Data Record – The Sequence
number allocated by kinesis , a partition
key a kind of routing code created by the
Data provider.
• Size up to 1 MB
• Firehose manages service for delivering
real-time streaming data to destination.
• Stream Analysis process , compute and
aggregate streaming data using standard sql.
This service simplifies streaming time series
feeding data to real-time dashboard and near
real-time dashboards.
AWS Kinesis

AWS GLUE vs. AWS Lambda
• ETL: Allows to split up jobs into small pieces that
can be handle Aschronoulsy.
• Integrate well with Amazon components.
• Data Processing : Execute code in response to the
trigger as such as change in data (S3 bucket) ,
Update to table in DynamoDB,Code Commit,
Cloudwatch (response to alarm), to trigger as
response to Inbound HTTP request(API Gateway)
and as a response to inbound message(SNS ,
Kinesis) or schedule event. AWS mobile SDK
conveniently support invocation of lambda
functions from mobile apps.
• Components : Lambda functions can be monitor
by using Amazon CloudWatch.
• ETL : Better for processing large batches of data
at once.
• Integrate with tool like Apache Spark well as it
primarily built on to execute framework.
• Data Processing : Automatically crawls your
data sources , identifies data formats and then
suggest schemas and transformations. Designed
to simply moving and transforming your dataset
for analysis.
• Components : DataCatalog,Crawler,ETL Jobs.
Monitor by Amazon CloudWatch

AWS Kinesis Dynamo DB
• Partition Key: The Partition Key is a string that is hashed to
determine which shard the record goes to. For instance,
given record r = {name: ‘Jane’, city: ‘New York’} one can, for
example, specify the Partition Key as r[“city”]which will
send all the records with the same city to the same shard.
• Dynamo DB AWS Cloud No SQL Database and rest of
NoSQL Database Shown as below.
• Kinesis Used for Streaming real-time NoSQL Data.
• Dynamo DB Used for real-time transaction streaming to
Store and it’s NOSQL Store.
• Amazon Redshift : Used for Big Data Analysis. A Columnar
Store
A stream A queue for incoming data to reside in. Streams
are labeled by a string. For example, Amazon might have
an “Orders” stream, a “Customer-Review” stream, and
so on.
A shard A stream can be composed of one or more shards.
One shard can read data at a rate of up to 2 MB/sec and
can write up to 1,000 records/sec up to a max of 1
MB/sec. A user should specify the number of shards that
coincides with the amount of data expected to be
present in their system. Pricing of Kinesis streams is
done on a per/shard basis
Producer A producer is a source of data, typically generated
external to your system in real-world applications (e.g.
user click data).
Consumer Once the data is placed in a stream, it can be
processed and stored somewhere (e.g. on HDFS or a
database). Anything that reads in data from a stream is
said to be a consumer.

AWS GLUE
Data lineage,Meta Data Description,Governace
• Data Stewardhip Role Curate data sources
Enrich technical meta-data and use tag to
indicate the quality.
• Below depicts what is Dynamic Catalog.
• AWS Glue discovery data and stores the
associated metadata in AWS Glue Data
Catalog.
• A fully manage ETL make the customer to
prepare and load data for analytics with ETL
jobs on few clicks.
• Data Virtualization : An Approach to data
management allows an application and
manipulate data without requiring technical
details also known by its format at the source
where it is located.
• Data Lineage : Defines data life cycle on how
the data moves over a time period from its orgin
and its get benefited by the visibility into
analytics pipeline allows us to trace the error
back to the sources.
Statistics %
• 76 % of company executive consider
information as mission critical.
• 60 % felt time constriant involved and lack of
knowledge of information on how to find the
right information which preventing the
associates what they want to need to know.
• Preliminary research found approx 35% to 50%
of information available yet centralized Indexed.
What AWS Glue really does with Data Catalog : Once cataloged, your data is immediately searchable, queryable, and available for ETL.

Amazon SNS vs. Amazon SQS
• Amazon SQS : - A Message Queue Service used
by distributed application to exchange message
through a polling model , used to decouple sending
and receiving components.
• Consumers pull message from SQS.
• Follows Jobs framework , Once the Jobs are
submitted to SQS, the consumer at other endpoint
process the jobs Asychronously. Depends on increase
in job frequency directly proportional to no: of
consumers for parallel processing.
• SQS stores message in distributed storage across all
availability zones with size up to 256 KB, can stores
unlimited #’s message across unlimited #’s of
Queues.
• SQS priced as function of #’s of request and transfer
that you made.
• Amazon SNS :- Allows Application to send
time-critical message to multiple subscriber which
prevents the periodically check or poll for updates.
• SNS pushes message to consumers by Push
mechanism.
• For e.g. An User uploading image to S3 , a
thumbnail along with a response email . In this way
S3 can send notification to SNS Topic and for e.g. if
3 potential customer who are tied to SNS Topic do
the following
1st
one watermark the image.
2nd
one create a thumbnail.
3rd
one Send a thank you email.
• In a Nutshell all of the 3 person receive same
image and doing their corresponding process in
parallel.
•
SQS Offers Serverless Queue no need to pay for Infrastructure , SNS Notify the service when a message arrives.
Relevant topics on Amazon SQS vs. Active MQ :-
https://thedulinreport.com/2015/09/05/top-ten-differences-between-activemq-and-amazon-sqs/

Amazon SNS vs. Amazon SQS
• Amazon SQS : - A Message Queue Service used
by distributed application to exchange message
through a polling model , used to decouple sending
and receiving components.
• Consumers pull message from SQS.
• Follows Jobs framework , Once the Jobs are
submitted to SQS, the consumer at other endpoint
process the jobs Asychronously. Depends on increase
in job frequency directly proportional to no: of
consumers for parallel processing.
• SQS stores message in distributed storage across all
availability zones with size up to 256 KB, can stores
unlimited #’s message across unlimited #’s of
Queues.
• SQS priced as function of #’s of request and transfer
that you made.
• https://thedulinreport.com/2015/09/05/top-ten-differenc
es-between-activemq-and-amazon-sqs/
• Amazon SNS :- Allows Application to send
time-critical message to multiple subscriber which
prevents the periodically check or poll for updates.
• SNS pushes message to consumers by Push
mechanism.
• For e.g. An User uploading image to S3 , a thumbnail
along with a response email . In this way S3 can send
notification to SNS Topic and for e.g. if 3 potential
customer who are tied to SNS Topic do the following
1st
one watermark the image.
2nd
one create a thumbnail.
3rd
one Send a thank you email.
• In a Nutshell all of the 3 person receive same image and
doing their corresponding process in parallel.
•
A subscriber is : A person who pays to receive a product or service, for example a magazine or website.
An Application : A computer software package that performs a specific function directly for an end user
A Push : When you push something, you use force to make it move away from you or away from its.
A Polling : In electronic communication, 'polling' is the continuous checking of other programs or devices by one program or
device to see what state they are in, usually to see whether they are still connected or want to communicate.

Natural Query Language
Natural Query Language : It’s an Interaction
between computer and human (natural)
language.
NLP are cater by the below types
1.Natural Language Processing.
2.Natural Language Understanding.
3.Automatic Speech Recognition.
LexNLP Use Cases
NLP – A Part of AI

The Goal of NLP is to fill the gap between
human and what the computer understand
Linguistic Analysis Consist of
1.Syntax
2.Semantics
3.Pragmatics
Syntax will tells us what part of given text is
grammatically true
Semantics will provide the meaning of the
given text.
Pragmatics define the purpose of the text
The mechanism of NLP involves two process
1.Natural Language Understanding.
2.Natural Language Generation.
Natural Language Understanding help us to
understand the meaning of a given text.
This also tries to understand and identify the
ambiguity present in natural language.
The Meaning of each word understood by using
lexicons(vocabulary) and a set of grammatical
rule. In addition to that there might be multiple
words having the same meaning(synonyms) and
words may have also more than one meaning
(polysemy).
List of Ambiguity :
Lexical : Words have multiple meaning
Syntactic : Sentences having multiple parse trees.
Semantic : Sentences having multiple meanings.
Anaphoric : Phrase/word which is previously
mentioned but having different meaning.
Natural Language Generation is a process of
automatically producing text from structure
data in an readable format with meaningful
phrases and sentences and cater by 3 stages.
1. Text Planning – Ordering of basic content in
structure data
2. Sentence Planning – Sentences are combined
from structure data to represent flow of
information.

Realization : Grammatically Correct Sentences are produced finally to represent text
Difference between NLP and Text Mining/Text Analytics :
NLP is responsible for understanding meaning and structure of given text whereas Text Mining/Text
Analytics is a process extracting hidden information inside text data through pattern recognition.
Popular Application of NLP :
1. Speech recognition
system
2. Question answering
system
3. Translation from one
specific language to
another specific
language
4. Text summarization
5. Sentiment analysis
6. Template-based chat
bots
7. Text classification
8. Topic segmentation
Popular Application of Text Mining :
1. Contextual Advertising
2. Content enrichment
3. Social media data analysis
4. Spam filtering
5. Fraud detection through claims investigation
TEXT Mining

NLP Process : TEXT Mining Process :
Click below
image

High Level Text Mining Process

Amazon Deep Learning Services
The Three Deep Learning Services in AWS:
1.Lex : e.g. Chatbot - ASR and NLU
2.Polly : e.g. TTS – Text to Speech
3.Rekognition : e.g. Face Recognition
Rekognition
Polly
Lex

AWS Textract
An AWS Service for OCR
Amazon Textract is a new Optical Character Recognition (OCR) service from Amazon that
allows customers to extract metadata across all types of documents.

Amazon Lex with Chatbot Vocabulary
Building Conversational interface into any application using voice and text
Chatbot Vocabulary with AWS LEX
1.Utterance
2.Entity
3.Intent
NLP : NLP examines an utterance and extracts
the intent and entities. NLP software includes
Amazon Lex, Facebook’s Wit.ai, and
Microsoft’s LUIS
4. Broadcast: A broadcast is a message sent
proactively to users. It is not a response to
user input.
5. Channel : Channels are the medium for
chatbot conversations. Examples of channels
include Facebook Messenger, Skype, Slack
and SMS. Email and web chat windows are
also mediums.(Lex as channel)
Conversational UI : User interfaces based on human
speech, either written or spoken. Conversational
UIs don’t use buttons, links or other graphical
elements.(Lex – AWS Service)
Pilot : The stage of development where the chatbot is
deployed to a small group of users for testing.
Proof-of-concept (POC): The stage of development
where the chatbot functions properly so long as
the input is artificially constrained.
Response: Anything bot says in response to user
input

Finding Insight and Relationship in Text
Insight in Text with Amazon AWS Services :
1.Transcriptions of Audio
2.Processing of Text
3.Text Analysis

Amazon Polly - An App That will Talk
from text to speech (integration with Node.js)
Amazon Polly : A Service that convert text into
lifelike speech. We can able to select up to 47
lifelike voices across 24 languages.
Creating Generic polly.js
Node Js
Run Index.js for computer to talk

Amazon Comprehend
https://activewizards.com/blog/comparison-of-the-most-useful-text-processing-apis/
Amazon Comprehend : is Natural Language
Processing Service that uses Machine Learning
to find meaning and insight into text.
This help us to identify language, extract
phrases, brands or events and understand
sentiments about product or services.
Most Common Use case :.

Amazon Transcribe Audio to Text Conversion in AWS
Amazon Transcribe : Transcribe recognizes
speech from audio files and covert it to text
A Sample scenario, the ability to differentiate
multiple speaker in the audio provide more
intelligible (“Who speak when”) and custom
vocabulary to improve the accuracy of speech
recognition for the product names.
APIBiology meant for vocabulary in the call to
synthesis speech API will help us to update
the vocabulary programmatically based on
errors.
Custom Vocabulary

ARRIA Natural Language Generation

ARRIA : Delivers patented natural language
generation (NLG) technologies that
replicate the way a human expert analyzes
data to pick out the important information,
and then generate natural language to
describe it.
Ref : https://youtu.be/Y5frfUzWAJI?t=105
Configuring and Generating a Narrative :
1. You need Tableau Desktop
2. ARRIA .TREX file to be download the
extension
(https://samples.arria.com/v2-tableau#e
xtension).
3. Step 1: Sample Tableau Dashboard.

3 Benefits Using ARRIA :
Out-of-the-box insights : This option lets you
tell the extension what it needs to know to
automatically generate insights from your
dataset. In the extension’s UI, this is
called Configure Narrative.
Extendable insights : In this option, you are able
to quickly customize the out-of-the-box
narratives by revising narrative scripts in NLG
Studio.
Programmable narratives : This option gives
you the most flexibility, and enables you to
create your own custom narratives for specific
use cases, with the full programmability of NLG
Studio. In the extension’s UI, this is called Create
Custom Narrative.
Extension URL :
https://samples.arria.com/v2-tableau/#exten
sion
Few Points on ARRIA Make the Difference :
• A Data Set cannot explain itself
• A Chart may represent but it will never
explain
• But ARRIA can provide the Explanation.
• To Provide a Narrative Extension to any data
service , developers will need to use a Natural
Language Service.
Below link depicts about importance of NLG :
https://searcherp.techtarget.com/feature/Natura
l-language-generation-software-turns-data-in
to-plain-English?utm_campaign=sfa_bizapps3
&utm_medium=social&utm_source=twitter&
utm_content=1483973376
NLG Projects for Reference :
http://mcs.open.ac.uk/nlg/#pastprojects
https://ehudreiter.com/2018/01/16/learn-about-
nlg/

Ref : https://samples.arria.com/v2-tableau-showcase/

Example of Natural Language Generation:
• Textual weather forecasts from numerical weather prediction models.
• Summary for patient from electronic patient record.
• Financial reports from finance spreadsheets.
BABY TALK
•Summarised clinical data about premature babies in neonatal ICU.
• Input: sensor data; records of actions and observations by medical staff.
• Output: multi-paragraph texts, summarised data for different audiences.
Analytics Insights on ARRIA 2018 :
https://cdn2.hubspot.net/hubfs/2047838/Mailings/2018-02-13%20Analytics%20Insight%2010
%20Best/AnalyticsInsight_Feb%202018.pdf
Competitor for ARRIA
https://sourceforge.net/software/compare/Arria-NLG-Studio-vs-Wordsmith/

Segmentation of AWS Components

AWS Services - Part 1

Recommended

More Related Content

Similar to AWS Services - Part 1 (8)

More from Sivakumar Ramar (7)

Recently uploaded (20)

AWS Services - Part 1