Federation University Australia at IIBIT (Sydney)

School of IT
Sydney Campus

Assignment Cover Page: Group Work

Course Number ITECH1103

Group Members
Last Name Given Names Student ID
SL
1 THIPPANI Priyanka 30371342

2. THUPALLI Sandeep 30356989

3. BHANDARI Sagar

4. AHMED Shoebullah Khan

COURSE NUMBER AND NAME: ITECH1103 Big data and analytics

PROGRAM OF STUDY: Master of Information technology

TUTORIAL GROUP: Lab 1 DAY/ TIME: Friday (10:30 – 12:30)

LECTURER: Md Monir Hossain

TUTOR (if applicable): Md Monir Hossain

TITLE OF ASSIGNMENT: Group Assignment (SAS visual Analytics)

WORD LENGTH: DUE DATE: 7-06-2019 DATE SUBMITTED: 7-06-2019

DECLARATION:

We have kept a copy of this assignment/work so that we can produce it if the original is lost or damaged. We hereby
certify that no part of this assignment/work has been copied from any other student’s present or previous, published
or unpublished, professional or amateur work or from any other source except where due acknowledgement is made
in the assignment. We further certify that no part of this assignment/work has been written/produced for us by any
other person except where such collaboration has been authorized by the unit/subject lecturer/tutor concerned

Priyanka, sagar, sandeep, shoeb

Signature of Student(s)

Note: It is necessary to sign the above declaration. A lecturer/tutor or an examiner reserves the right not to mark this
assignment/work if the declaration has not been duly signed.
Task 1

Data set:
Data sets are the files that hold the measurements, counts, and categorizations collected from
individuals and objects. A data set is a file stored in a library that creates and processes. Data set
contains number of data values that are organized as a table of observations (rows) and variables
(columns) that can be processed. The data set contains descriptor information such as data types
and lengths of the variables, which are used to create the data.

Project:
A project is a record of the data sets that can be opened, that can be run, the results that can be
produced, and it explains clearly about the relationship of the objects. The project which contains
the information is called the project file. It can be saved and copied.

Benefits of using SAS visual Analytics:

Using SAS Visual Analytics, user can enlarge the analytic power of the data, explore new data
sources, investigate them, and creates a clear explanation of the relevant patterns. The users can
share those relevant patterns in reports. By following the patterns the output can be well-defined
up-front and we can get what you are looking for and what we need to bring. However, data
discovery suggest you to erect the data, is characteristics, and its relationships. Then, when
proper visualizations are created, we can associate those visualizations into reports.

SAS Visual Analytics give the users with the following benefits:

 Enables user to apply power of SAS to the large amount of the data .
 Allow the users to visually explore the data, depends upon the measures, and at
incredibly fast speeds.
 Enable users to create the powerful Statistical data models easily.
 Enables users to create the reports by using tables and graphs.
 The graphs can be customized depends upon the data what you requires and the unwanted
data can be deleted easily.
 Enables users to share the data with anyone.
Justification of SAS successful in other similar projects:
Justification 1:
The business metrics in real time allows a company to understand and reply to ever-changing
customer demands. In reality, though, obtaining such metrics in real time isn't constantly easy.
However, SAS Australia and New Zealand Technical Support solved that hassle through the
usage of SAS Visual Analytics to develop a 16-display command center within the Sydney
office. Using this center to offer real-time facts allows the Sydney office to reply to client needs
throughout the complete South Asia region.

Figure 1: The SAS Architecture for the Command Center

After ANZ Technical Support scoped out the records necessities for the command center, their
subsequent step was to layout and builds the surroundings. The group determined to position the
ETL processing and SAS Visual Analytics Server onto separate servers. This layout simplified
the function of every server and made it simpler to preserve the environment and to troubleshoot
any technical issues encountered throughout the layout phase or after the assignment went stay.

Conclusion:

SAS has leveraged a breadth of SAS technologies to better understand the needs of its clients, to
enhance efficiencies, and to in the end drive consumer satisfaction via SAS Visual Analytics in
the command center. The ideas and strategies described in this paper provide a blueprint for
growing and constructing a command center.
Justification 2:
Using Excel as an analytics tool is pervasive and nearly inescapable in lots of company
environments. While it could be a super device for plenty fundamental ad hoc analyses, it can
easily morph into an unscaleable, error prone, and undocumented reporting answer with
substantial barriers on end-consumer functionality. When this takes place, SAS Visual Analytics
can provide a new and progressed reporting solution that addresses many of Excel's pitfalls. This
project uses an actual-world example to illustrate the blessings of moving Excel-based reporting
to SAS Visual Analytics.

This project discusses several key benefits of the SAS Visual Analytics solution together with:
consolidating a plethora of Excel files right into a small list of dynamic reports, greatly reducing
the hazard of human mistakes in reports, decreasing the quantity of resource hours needed to
hold and decorate reports, and providing end-users with the advanced functionality they want for
powerful decision making.

Figure 2: Consolidating numerous excel files into a small list of dynamic reports

The team commenced this project with an intensive assessment of the legacy reporting procedure
including an evaluation of facts assets, documentation of business policies, cataloguing of reports, and
identity of all the manual steps necessary for document creation.

Conclusion:
The cal center reporting solution from Excel to SAS Visual Analytics reduced the variety of
stories by using extra than 20 and provided end-users with added functionality. The simplicity of
SAS Visual Analytics allowed for the rapid development of latest reviews. Significant benefits
were achieved via the automation of manual steps, standardization of business rules and
regulations, and the capability to redirect analyst time away from guide manual maintenance to
more cost-added initiatives. These benefits make SAS Visual Analytics a brilliant Excel
opportunity for organizations in comparable situations.
Justification 3:
At the University of Central Florida (UCF) recently invested in SAS Visual Analytics, along
with the updated SAS Business Intelligence platform a mission that took over a year to be
finished. This mission became undertaken to offer the customers the best and maximum up to
date equipment to be had. This project introduces the SAS Visual Analytics surroundings at UCF
and includes projects created the use of this product. It answers why they selected SAS Visual
Analytics for development over other SAS packages. It explains the technical surroundings for
our non-disbursed SAS Visual Analytics: RAM, servers, benchmarking, sizing, and scaling.

Figure 3: Targeted Performance Measure Dashboard

The subsequent step became to create dynamic reviews that might allow user interactivity and to
provide analytical visualizations of the information. One such file this is main development is the
“Targeted Performance Measure Dashboard”. This report is staged to serve as UCF’s
government dashboard displaying information on five selected metrics at the university stage
with the ability to drill into schools and departments. This changed into a capability that we
couldn't easily accomplish without using a couple of reviews with a couple of maps.

Conclusion:
Performing SAS upgrades or installations may become a challenging task. There is an awful lot
practise, studies, and frustrating tries concerned in the system. SAS platform versatility can be
adapted to clearly any device, accordingly making its degree of personalization and tweaking a
very problematic procedure. As a result, many hours must be invested communicating with SAS
to remedy ‘issues’ which might be no greater than mere unique setup steps or very particular
situations, now not necessarily disclosed with general set up techniques due to the style of
‘feasible’ eventualities.
Justification 4:
The Institute for Advanced Analytics struggled to provide student computing environments able
to studying increasingly more large records units for its Master of Science in Analytics program.
For the quick-paced practicum, the center-piece of the curriculum, waiting 24 hours for a FREQ
procedure to complete was unacceptable. Practicum proposals from industry had been pared
down (or grew to become down) due to the fact the data sets had been too big, depriving students
of exciting and applicable studying reports.

By augmenting the practicum architecture with an 18-node computing cluster running SAS Grid
Manager, SAS Visual Analytics, and the modern High-Performance Analytics methods, we were
capable of dramatically growth overall performance and begin accepting terabyte-scale
practicum proposals from industry. The project tells about the benefits and lessons learned
through including those SAS products to our analytics degree program including functionality
versus complexity tradeoffs, and the kingdom of our cutting-edge abilities and barriers with this
structure.

Figure 4: Software architecture of third generation (current) practicum grid

Conclusion:
By augmenting the practicum structure with an 18-node grid running a group of SAS software,
we substantially multiplied computing resources for the student teams by way of factors.
Empowered by means of this new hardware, we installed SAS Grid Manager, SAS Visual
Analytics and High-Performance Analytics approaches on the new grid. The integration of the
brand new grid with the present student team servers turned into complicated and tough.
Running multiple record structures across 2 operating structures and dealing with 24 separate
safety domains turned into a amazing undertaking to manage and of which to teach users.
However, the blessings had been very valuable to them, as the brand new system allowed for
terribly big present day practicum tasks that might in any other case be disregarded due to
prohibitive “large data” demanding situations.
Justification 5:
SAS Visual Analytics has the skills to discover traits, see relationships, and percentage the
consequences along with your records clients. This project explains a case study making use of
the talents of SAS Visual Analytics to NCAA Division I college soccer information from 2005
through 2014. It follows the manner from reading uncooked comma-separated values (CSV) files
through processing that statistics into SAS facts units, doing facts enrichment, and sooner or later
loading the information into in-memory SAS LASR tables. The case examine then demonstrates
using SAS Visual Analytics to discover specified play-by means of-play information to find out
trends and relationships, as well as to analyse team inclinations to broaden sport-time techniques.

Figure 5: SAS Visual Analytics Explorer

The explorer allows us to study the CSV report containing the person play facts for each game.
Our exploration of the information gives us insight as to the values of character columns, how
the values are disbursed, and how analysis-prepared the fact is. This facilitates us know what
statistics is already to be had within the uncooked statistics, and what we may additionally want
to derive from the uncooked information, to convert it into a presentation-geared up kingdom.

Conclusion:
This project, which commenced as a workout to the test SAS Visual Analytics, to examine the
functions and abilities of the product. This additionally provided an automobile to do a case look
at of one technique to designing and implementing an answer that involves several extra SAS
services. We have visible a number of the skills furnished via SAS Visual Analytics, but there
are many extra to be had.
2 - Dashboard/Reports

1. Create a data dictionary for the data source for use by the group.

A) In data menu bar we have data properties, measure details and data source details.

In BIRDSTRIKE_DATA we have category variable of 24 and measure variable of 13.

Data properties:
Measure data:

Data properties:
Data source details:
Measure details:

2. How many different species (distinct) have been involved in bird strikes?
A) Different species have been involved in the bird strike. The below bar graph shows the
species of birds involved in bird strike.
There are 609 distinct values of species according to the given data.
3. Where the data has been recorded, what percentage of the time was the pilot warned of birds
or wildlife?
A) The blow bar graph shows the percentage of pilot warned of birds or wildlife. There are
three distinct values for this data.
4. What are the top five airline/operators involved in bird strikes?
A) The top five airline/operators involved in bird strikes are Business, Military, Southwest
Airlines, United Airlines and unknown(Which is not mentioned the exact name of the
airline in data)

5. What proportion (percentage) of bird strikes involved medium and large wildlife?
A) The percentage of bird strikes involved in medium is 85.75% and large wildlife is
14.25%.

6. In which phase of flight would the aircraft be most likely to experience a bird strike?
 A) In the below bar graph the missing data of phase of flight the aircraft be mostly to
experience a bird strike.

7. Has a bird strike ever occurred whilst the aircraft was parked? If so, in how many instances?

A) In phase of flight data only we can see the aircraft parked. It is shown that from below bar
graph we can tell there are 25 aircraft has been parked.
8. Is a bird strike more likely to happen when there is no cloud as compared with
overcast conditions?
A) Yes, bird strike is more likely to happen when there is no clouds as
compared with overcast condition.
9. In what precipitation conditions is a bird strike most likely to happen?
10.Which species of wildlife had the largest total cost? How much was this?

A)The canada goose of wildlife has the largest total cost of 80,080,800 and next
followed by white – tailed deer and bald eagle.
11.For the most costly species of wildlife, which phase of flight was the most
common for a bird strike?
A) The most costly species in wildlife is Canada goose, climb is the phase of
flight in which the most common for a bird strike.
12. For the species Bald Eagle and Canadian goose, examine the time of day that
bird strikes occur to see if there are any differences between the two species.
Hint: you will need to use a filter on the species of wildlife and also use two
categories.

A)
13.Investigate using a crosstab for the data values Pilot warned of bird strike,
Altitude bin, and the total cost.
A) We need to select the crosstab and select the character variables of pilot
warned by bird strike in column and altitude bin in rows. Total cost is a
measure value in measure.
14.Create a geomap of the bird strike data.
A) For geomap of bird strike data we need to select geomap option it will
display world map. I took a screen shot of Australian map for bird strike.
15.Using the sum of the number of records, investigate the bird strike count by
airport. Hint: remove any missing data and use a sort.
A) In this the sum of records and bird strike count is the measure values and
airport names are the variable character. By using filter we have to remove
the missing data. In this way we got the below bar graph.
16.Investigate the bird strike cost by airport. Hint: remove any missing data and
use a sort on the cost.
A) The bar graph gives the bird strike cost by the airport.
17.Compare the answers of the two questions above. What does this indicate?
18.Perform a cluster analysis on the two variables time out of service and total
cost. Use the cluster matrix graph to describe how the four clusters differ
from one another in terms of the two variables used
A) For this we need to select the cluster matrix graph for two different variables
of Time out of service and total cost. Hence, we get the below graphs
 3 - Additional Visualizations

1) How many types of Aircraft has been involved in bird strike

A) There are 4 distinct values for aircraft. In which they are involved for the
bird strike. More percentage is of airplanes.
2) For the species sparrow, crow and pigeon, examine the time of day that bird
strikes occur to see if there are any differences between the three species.
Hint: you will need to use a filter on the species of wildlife and also use two
categories.
3) What sizes of wildlife have been involved in bird strike
A) The sizes of wildlife involved in bird strike are small, medium and large.
4) What is the impact of flight on bird strike
5) How many number of engines are there for aircrafts
A)
6) In what sky conditions is a bird strike most likely to happen
A)
References:
Senge, Peter M. 1990. The Fifth Discipline. The Art & Practice of the Learning Organization. New York,
NY: Doubleday

Vitron, Christine. Holman, James. “Considerations for Adding SAS® Visual Analytics to an Existing SAS®
Business Intelligence Deployment.” SAS Global Forum 2014. March 23-26, 2014. Washington, DC.
Available at: http://support.sas.com/resources/papers/proceedings14/SAS146-2014.pdf