Scribd
Upload
69 views

Big Data Requirement Gathering

Amazon uses big data analytics to better understand customer preferences and predict what customers will buy. It collects and analyzes data from customer purchases, searches, and other interactions to build a robust understanding of individual customers and make personalized recommendations. This helps customers find relevant products more easily and improves Amazon's sales and profits. Amazon employs technologies like Hadoop and data warehousing to manage the huge volumes of customer data it collects and gain valuable business insights through data visualization and analysis.

Uploaded by

inder saini
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
69 views

Big Data Requirement Gathering

Amazon uses big data analytics to better understand customer preferences and predict what customers will buy. It collects and analyzes data from customer purchases, searches, and other interactions to build a robust understanding of individual customers and make personalized recommendations. This helps customers find relevant products more easily and improves Amazon's sales and profits. Amazon employs technologies like Hadoop and data warehousing to manage the huge volumes of customer data it collects and gain valuable business insights through data visualization and analysis.

Uploaded by

inder saini
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 11

N01346254 Inderjit Singh

1. Overall Project Description


a. Use case Title
Amazon Big data Analytics and sales prediction

b. Use case Description


Amazon has adopted the model which can be significantly named as
“everything under one roof”. Customers get huge variety of options and
they become data rich with tons of options. But, they end with having
less idea about what would be the best option available for them.
To prevent this, Amazon uses the data collected from the consumers
while building up their furnished engine for data maintenance. Amazon
gets the requirements from the buyers and predicts what they buy. Thus,
Amazon streamlines the process of recommending variety and high
volume of products and makes the customers search filtered.
c. Domain- Vertical
 Security
 Storage
 Health care products
 Gaming domains
 Manufacturing resources
 Industrial IoT
 Media and Entertainment
 Telecommunications
 House Hold utilities

d. Applications
 Application Servers
Amazon has lightweight servers with high performance which
provides high volume of connections and maintain reliability and
security.
 Application Stacks
Amazon has variety of application stacks and there is bunch of teams
to manage it. MySQL and DynamoDB are most used resources in
application stacks. Web based Applications are commonly
recommended by the customers thus mostly application stacks of
amazon include angular JS and jQuery along with hive and Hadoop
architecture for data maintenance.
 Monitoring
Amazon monitors the services with the help of DevOps , site
reliability engineers and IT managers. CloudWatch provides the
resources to monitor the applications and unique view of
operational data. Data collection is done in the form of logs, events
that run on-premises servers to troubleshoot issues and to maintain
applications to run smoothly.
 Log analysis
Amazon maintains log analytics which includes the most searched
data , analyzing and visualizing the purchases and the orders
demanded by consumers in order to get operational insights.
Amazon has a fully managed service which gathers the logs and
metrics across the web site to provide visibility of trends in the
applications to provide flexibility , scalability and thus profitability
for log analytics.

 Intelligence Automation
Amazon has artificial intelligence services by which it gets the
most demanded services and orders asked by customers and it
automatically detects the searches of customers and their history of
purchases. Thus, it becomes more effective for customers because
they are presented with the related products they will actually go
for.
 Price Optimization
Amazon interacts with customers by advertisements, by publishing
dummy products and by conducting surveys and many other
strategies. Thus, Amazon manage the costs of products by
surveying the end users by observing purchases, current trends ,
quantities of sales and log analytics in order to get price
optimization.

 Customer’s View
Amazon maintains customer view by providing various services
like Ordering options, Delivery options, physical options and
return options.
1. Ordering options provide one-click option and Dash button for
customers. With one-click option, Amazon allows the
customers to buy products by pre-set options (credit cards and
shipping address). Dash- button options provides the customers
with more detailed steps and also the return related options.
2. Delivery options provides different shipping details. Most
recently, Amazon has adopted “self service kiosks” where users
can access the packages which can be easily delivered for easy
pick ups.
3. Physical options of Amazon provides “Amazon Fresh” with
delivery of groceries at the user’s home and also other key
market access to the customers.
4. Return options provides customers with return labels and drop
off locations. It is very important for a retailer to know how the
consumers are willing to go for returns.

 Data Warehouse offloading


Amazon has adopted Big data technology like Hadoop along with
data warehouse to ease the management of growing data in order
to reduce overall costs. This has resulted in solving the problems
which result from accessing high volumes of data in the
warehouse. Customers place thousands of orders every minute, so
to maintain those requests Hadoop architecture comes into play.

e. Project goals and objectives


 To help Amazon to attract the customers with most efficient
services.
 To help Amazon to manage the high volumes of requests.
 To help Amazon to provide IT support for the maintenance of
items and various shipping criteria.
 To maintain overall price optimization.
 To incorporate more and more use of Artificial intelligence and
Hadoop file systems for the overall growth of premises.
2. Big Data Characteristics
a. Data sources
Loyalty cards, Debit cards, Credit cards
By providing customers with deals and rewarding them with discounts are
easy way to collect data on larger scale. Also, debit and credit cards are
valuable data sources. These cab also be used for tracking purposes.
Web Logs
This technique helps the retailers to get to know how many customers are
accessing the online web site of Amazon. This helps to get the most visited
pages as data sources.
Financial Invoices
The invoices also act as data source. When the customer places the order or
buys some items, then the invoice history is maintained to track the purchase
any time.
b. Data destination
Data is being analyzed to provide meaning to it. Amazon has data analyst
which may be an IT professional who uses HDFS to process the data in
Hive. Thus, Big data support is provided to handle huge volumes of
customer requests, item histories, purchases to provide data integrity and
thus data drives the change in overall performance of system.
c. Data volume
Data volume means the amount of data being processed.
At macro level:
 Amazon in the peak season deals with huge volume of data.
 Data comes from customer’s requests for items.
 Data comes from orders placed by customers.
 Millions of customers access the website online in minute.
 Thousands of transactions take place.
 100’s of petabytes of data is being stored in HDFS architecture.
At micro level
 Usually, when there are less sales, only hundreds of transactions take
place in a minute.
 Data volume is less this time.
d. Data Velocity
 Thousands of surveys being done in a day.
 Thousands of transactions are done in a minute
 Millions of people visit the web site every minute.
 Transfer of products and shipments in huge amounts every minute.

e. Data Variety
 Different Databases but with unique model(.sql,.mdf formats)
 Different kinds of file formats being supported by file system.(reports
in .xls, .doc ,jpg ,.png)
 Huge amount of unstructured data being generated and processed.
 Audio(.mp3), video(.mp4), image streams of data being processed.
3. Final Stage
 Data visualization is being done in the form of graphs and charts.
 Data visualization is done by Analyst to make decisions.

Amazon net revenue prediction graph:


Key Requirements for data visualization in Amazon:
Net Promoter Score(NPS) : It means how likely is the customer to
recommend for the item or services to the peers.
Customer Profitability Score: How much profit customer brings to the
business. This is usually mentioned in graphs to be accessed by analyst.
Conversion rate: What is the conversion rate of the current anaylysis to the
expected decisions with profitability.
Relative market share: What is the position of the company with respect
other companies in market.
Net Profit Margin : The percentage of net revenues of the company.

You might also like