MITS6002 Assignment 3: Regression and Classification Analysis

Desklib provides solved assignments and past papers to help students succeed.

MITS6002
ASSIGNMENT 3

Contents
Q1....................................................................................................................................................3
Answer i:......................................................................................................................................3
Answer ii:.....................................................................................................................................3
Answer iii:...................................................................................................................................4
Answer iv:....................................................................................................................................4
Q2....................................................................................................................................................5
Answer i:......................................................................................................................................5
Answer ii:.....................................................................................................................................5
Answer iii:...................................................................................................................................6
Answer iv:....................................................................................................................................6
Answer v:.....................................................................................................................................7
Answer vi:....................................................................................................................................8
Q3....................................................................................................................................................9
Answer i:......................................................................................................................................9
Answer ii:.....................................................................................................................................9
Answer iii:.................................................................................................................................10
Answer iv:..................................................................................................................................11
References......................................................................................................................................13

Q1. Carefully Read the “CommBank Retail Business Insights Report FY18” provided with this
as an attachment and answer the below questions.
i. Comment on the insights report based on the overall features; including the quality of
visualisations, presentability, and the information provided.
Answer i:
Here, in this insights report for the CommonBank Retail Business has been explained which was
done on the basis of the survey of almost 2,473 business owners, different decision makers as
well as managers and 16 qualitative interviews taken from different highly skilled professionals.
This report provides detailed information about the results of the survey, deep interviews and
others in order to show the benefits of innovation in a different business to enhance the growth as
well as the performance of the business in Australia region. All the data has been presented in the
best way along with better and attractive visualization to attract a large number of viewer and
readers towards the report. The overall insights report has been presented in short and simple
manner with easy to understand language for the viewer so that they can easily understand the
visual data presented in the report and use those results for the enhancement of their business.
ii. List the key information you derive from this insights report and explain how they
will be useful in decision making.
Answer ii:
In this insights report, different information has been shown as a key point for the complete
report in order provide sufficient information as well as data in visual form along with their
analysis to present the result of the different survey, interviews, contribution of decision makers,
etc. The main key information has been given below:
 Innovation performance in retail, online and multichannel business.
 The behavior of retailers towards innovation.
 Major drivers play an important role in innovation & improvement in the retail sector.
 Challenges related to innovation in the retail business.
 Major areas of investment.
 New technologies which need to be used in the retail sector.
 Estimated return on the basis of investment in the retail business.

All these points are key information which plays an important role in the decision making for
any business owner as well as other retail industries in order to implement innovation in their
business to enhance the performance as well as the growth of their business.
iii. Write an abstract (one paragraph) summarising the insights report.
Answer iii:
With the increasing competition in the retail business of Australia region, different business
owners started implementing new innovative ideas in their business in order to enhance their
business as well as to grow their business performance with the help of latest technologies and
resources which are available. As a result, multiple retail businesses got success in improving
their business as well as getting proper returns from their investment while different other
retailers are still suffering which need to be improved. This report has been developed and
presented using different data in the form of the visual presentation along with analysis of the
different survey, interviews, and decision maker’s contribution for those retailers who are still
suffering in order to grow and enhance their business with correct decision as well as investment.
iv. Suggest improvements to this insights report.
Answer iv:
Here, in this insights report, all the data has been presented with the help of bar graph on the
basis of overall result which was collected from different survey and interviews including 262
retailers from different sectors such as home, clothing, etc. on the behalf of Commonwealth
Bank. These data can be presented with more visual representation using different attractive
graphs instead on using only bar graph along with some examples, references, and images of the
top benefited retailers in order to build a better level of trust between report and the readers.

Q2. Regression analysis is a commonly used technique to find relationships among variables.
Answer the below questions based on regression analysis.
i. Provide an example where regression analysis can be effectively used.
Answer i:
Regression analysis is defined as a methodology for the analysis of functional relations in-
between different variables. For example, analysis of selling price for a house or flat on the basis
of different taxes as well as physical characteristics of those houses and flats which is done by
real estate agents (Chatterjee and Hadi, 2015).
One of the best examples where regression analysis is used effectively for the analysis of
feedback or survey forms collected from different customers and users on the basis of quality,
service, and satisfaction for the products and services they had purchased in order to enhance and
improve their services as well as product in future.
ii. Collect height and weight data from 10 friends/relatives of yours and complete the
below table. Every student in class should have a unique set of values.
Answer ii:
Name Height (in cm) Weight (in Kg)
Steven 144 65
Sam 148 60
Asher 154 55
Tom 158 62
Laurel 151 50
Marks 160 72
Jenifer 147 48
Samuel 157 58
David 149 75

William 153 85
iii. Draw a scatterplot based on above data. Based on your plot comment on the
relationship between height and weight.
Answer iii:
The scatterplot based on above data of height and weights for 10 friends has been shown below:
Figure 1: Scatterplot
From the above graph, it can be stated that when height taken in cm is increasing in the x-axis,
the weight of different friends taken in Kg placed in the y-axis is also increasing in the
scatterplot. This scatterplot has been developed using all the details of height and weight of 10
friends which have been inserted and the result was developed on the basis of the analysis of
inserted data.
iv. Compute the equation of the regression line.
Answer iv:
From the above values and their calculations, the equation of regression line is developed as
follow:

From the scatterplot;
Sum of X = 1521
Sum of Y = 630
Mean X = 152.1
Mean Y = 63
Sum of squares (SSX) = 244.9
Sum of products (SP) = 95
Using above values for the slope, we have equation of regression line;
Y = bX + a
Where, b = SP/ SSX = 95/244.9
= 0.38791
And, a = Mean Y – b* Mean X
= 63 - (0.39*152.1)
= 3.99837
Now, putting above values, we have;
Y = 0.38791X + 3.99837
v. Calculate the R2 value and comment on the goodness of the fit.
Answer v:
In the regression analysis, R square and goodness of fit lays an important role in the analysis of
data as well as decision making. Here, goodness of fit is defined as the squared value of R in
proportion of difference or variance in the value of Y calculated on the basis of variation of
values for the X. It helps in analysis of close data which is best fitted for the output or the value
of Y (Bartlett, 2014).
From above scatterplot, calculated value of R square for goodness of fit are as follow:

R2 (R square) = 0.03056
vi. Use an analytics tool of your choice to calculate the values for iv, and v. Compare
them with your answer.
Answer vi:
Here, an online tool named “GraphPad” has been used for the calculation of the results on the
above data for the regression analysis (Graphpad, 2018). All the results which have been
developed using the above tool have been given below:
Best-fit values
Slope 0.3879 ± 0.7725
Y-intercept 3.998 ± 117.6
X-intercept -10.31
1/Slope 2.578
Goodness of Fit
R square 0.03056
Sy.x 12.09
And,
Equation, Y = 0.3879*X + 3.998
From the above results, it can be concluded that both manually calculated results are very close
to the results calculated using calculator.

Q3 Classification and regression are commonly used processes in business analytics.
i. Briefly explain the difference between classification and prediction.
Answer i:
Classification and prediction are two important terms in data mining which plays an important
role in the analysis and calculations for different data in order to enhance the performance as
well as the profit of the organization. The major difference between classification and predictions
has been shown below:
Table 1: Table of difference between classification and prediction
Classification Predictions
Identification of class or category Identification of missing values
It belongs to the observation It belongs to the numerical data of an
observation
Accuracy of the calculation depends on the
correct identification of class
Accuracy depends on the best prediction
which was predicted
Construction of models done for the
identification of class
Construction of predicted values done by the
predictor.
Here, another name of the model is classifier Here, another name of the model is a
predictor
(Mandula, 2018)
ii. Give examples for classification methods you know.
Answer ii:
In data mining, there are different methods or algorithms available for the classification. Some of
the best-known examples of different classification methods are as follow:
Logical Regression: In this method of classification, binary output is developed depending upon
the input value i.e. 0 or 1, true or false and yes or no.

Decision tree: In this method of classification, complete data is divided into two or more than
two data sets depending upon the input data in order to make different groups with different
values differ from one other.
iii. The following diagram shows a neural network with one hidden layer.
Write down the algebraic equation for y1 in terms of input values i1,i2 and weights w. Briefly
explain how neural networks are used for classification.
Answer iii:
From above diagram,
Taking hidden layer as Layer 1, we have;
Z(1) = W(1)*X + b(1) ---------------- equation (i)
And, a(1) = Z(1) ----------------------------equation (ii)
Where, Z(1) = Output of Layer 1 in vector form
W(1) = Weight of neurons for the hidden layer in vector form (w1, w2, etc.)
X = Input which is i1 and i2 in vector form
b = Bias of hidden neurons in vector form which is b1 and b2
a(1) = Liner function in vector form
Now, taking Layer 2 which is output layer along with 2 as output from the 1 layer, we have;
Z(2) = W(2)*a(1) + b(2) ---------------equation (iii)

And, a(2) = Z(2) -----------------------------equation (iv)
Hence, using above equations, we have;
Z(2) = (W(2) * [W(1)X + b(1)]) + b(2) = [W(2) * W(1)] * X + [W(2)*b(1) + b(2)]
From above, we get;
W = [W(2) * W(1)]
And,
b = [W(2)*b(1) + b(2)]
Thus, the final result will be;
Linear function, Z(2) = W*X + b
In data analysis as well as computation, the neural network works same as the one which works
in the human body in which different neurons works on the basis of bias, the weight of the
human body and their respective functions. In the same manner, once weight and bias are
provided to the neuron network, final output calculated on the basis of input weight and bias
provided to the network along with the error at input and output of the network to produce the
final result (Woodford, 2019).
iv. Give at least three examples how clustering can be used in business analytics. In your
answer explain how each business case could be addressed using clustering.
Answer iv:
Clustering: It is defined as a process to develop different small groups of similar objects, data,
values and other from given a large amount of data, population or values (Vohra, 2018). Some of
the examples of clustering in business has been given below:
Example 1: In order to open emergency unit within a hospital to cover maximum regions where
the accident occurred, identification of the location of emergency units is difficult to calculate
which can be easy using clustering in order to cover all the critical or accident-prone areas by the
emergency unit of the hospital.

Example 2: To open delivery center for the Pizza in the city, different challenges occurred such
as identification of areas from where Pizza is being ordered mostly and frequently, the total
number of stores which needs to be opened and other. All these challenges can be simplified
using the clustering algorithm. The complete steps have been shown below in the image:
Figure 2: Example 2 explained in step by step process
Example 3: Classification of crime in any country based on the consumption of drugs such as
Heroin, Cocaine and other which is difficult for calculation manually but it can be identified in a
simple way using clustering algorithm in order to identify the major areas of crime, drugs used
and other data.