MITS6002 Business Analytics Assignment 3: Retail Innovation & Data

This assignment delves into key business analytics concepts, starting with an analysis of retail innovation based on a report encompassing data from over 2300 businesses and qualitative interviews. It explores the drivers and benefits of innovation for retailers, including investment areas. The assignment then transitions to regression analysis, demonstrating its application in measuring risk and identifying opportunities, using a provided dataset to calculate regression line equations and coefficients. Finally, it differentiates between classification and prediction, outlining various classification methods and their business applications, particularly focusing on clustering techniques for customer analysis, product analysis, and location-based investment strategies. Desklib offers a wide range of solved assignments and resources for students.

MITS6002 Business Analytics
Assignment 3

Q1.
i. This report is based on the information from more than 2300+ businesses, decision
makers and managers also, qualitative interviews from more than 15 personal
interviews. Both the information sources have information which provides both
qualitative and quantitative foundation. The retailers and business owner are the
subsets of a large dataset as we are using a data analytics tool to find which areas are
changing the most and by which factor. In this report, we will discuss the dynamics of
innovation which means determining what the major factors which impact the
retailers to adapt to new technology. Some of the questions in this report are,
1. Drivers of innovation and improvement for retailers
2. Perceived benefits of innovation and improvement for retailers
3. Innovation behaviors of retailers
This report also performs discussion and analysis on cases like Investment in
innovation and areas of investment.
ii. After reading out the provided report, we were able to understand the importance of
data analytics in the business sector. There are many facts, figures and statistics to
understand what change is happening due to which trend.
There are some insights like understanding the retail business and find out what are
the circumstance and factors which are making retailer to use more technology in
their business.
iii. This report includes statistics, facts, figures about how retailing business is evolving.
There are discussions about the impact of technology on the Australian retailing
business and what is the reason for business owners for using the latest technology in
their business.
iv. In terms of statistics, this report is well informed but to improve it a bit more, there
can also be word relation between facts and figures of the past compared to the future.
Also, there can also be a reference to some of the sources for these figures so that
anyone can get a better understanding of the in-depth report information.

Q2
i. Many businesses use regression analyses to measure the risk and future opportunities
to help the business in the most efficient way. For a decade, major companies like
Google, Amazon, IBM, etc. have been using technologies like data analysis and Big
data to uncover truths about efficiency and using those methods to reach a new level
of success. In the business sector of data science or regression analysis, people use
various devices and sensors to see beyond the displayed data. Regression analysis is
used to recognize patterns and find similarities to make decisions based on these
analyses. These technologies have very high precision in seeing data patterns to
predict the best and possible next step. As we know that, regression analysis is all
about data and in business we use where the data marks certain data point which we
analyze to determine the relationship between the data are given and prediction
results from that given data.
Not only that, regression analysis help business in understanding the patterns,
correction of errors, process optimization. All these factors are very important in
order to make any business successful and regression analysis ensures these processes
are running efficiently so that prediction results are viable and applicable for the
desired business.
ii. Height
(inches)
Weight (pounds)
76 225
75 195
72 180
82 231
69 185
74 190
75 228
71 200
75 230
72 175

iii.
Figure 1: Graph for the above data set
iv. Sample size: 10
Mean x (x̄): 74.1
Mean y (ȳ): 203.9
Intercept (a): -132.20814880423
Slope (b): 4.5358724534984
Regression line equation: y=4.5358724534984x-132.20814880423
v. R2= 0.516
vi. Estimated coefficients:
b_0 = -132.208148804251
b_1 = 4.535872453498665
R2=0.5169979931529012
This graph below is drawn using python, python is a very strong business analytics
tool. Many big companies like Google, Amazon, Genpact, etc uses python to analyze
data and find the relationship between current information and future predictions.

Comparing the values from two graphs and their values, it is usual that both graph
values are the same as well as the graph which means that analytics tools are also
based on the same concepts.
Q3.
Classification Prediction
In classification, a classifier also known as the
model is created to predict the label which is
categorized.
In prediction, we don’t get a yes or no answer
instead the result is a numerical value which
is predicted using various data analytics tools
in other words, a predictor is created to
predict ordered values.
The data in classification can only be
determined in two options, yes and no.
The result will always be a numerical value.
For example: If in a bank, the loan officer
needs to perform data analysis on the
customer on the basis to determine which are
risky or which are safe.
For example: In a company, manager wants to
predict how much a specific customer will
spend on their e-commerce website in the
upcoming quarter.
ii. The various classification methods used in data analysis:
a. Linear Classifiers: Logistic Regression, Naive Bayes Classifier
b. Support Vector Machines or SVM
c. Decision Trees
d. Neural Networks
e. Nearest Neighbour
iii. Equation
iv. Clustering in business analysis is used among various situations some of them are as follows,

1. In the world of e-commerce, clustering can be used for customer analysis, which
means determine which set of customers are most likely to become big sender in the
upcoming quarter.
2. Also, e-commerce business can use clustering, to analyze which products are getting
the most purchases, so that the company can predict which items must be kept in
abundance.
3. Clustering can be also helpful in case of finding which areas are showing the most
potential so that the company can invest its resources to get the maximum results
from most promising locations.