COIT20249 - Machine Learning for Merchandise Business: Report Analysis

This report, prepared by ABC consultancy for JD, investigates the potential of machine learning to enhance its merchandise business processes. It defines machine learning, differentiates it from artificial intelligence, and analyzes its impact across various industries, including automotive, finance, and transportation. The report provides a detailed application strategy for JD, focusing on price optimization, marketing, and consumer management, outlining the advantages and disadvantages of implementation. It addresses ethical and technical issues, such as job displacement and infrastructure requirements, before offering recommendations for effective machine learning integration. The report concludes with strategies to ensure effective application of machine learning in merchandise business, emphasizing the need for data collection, algorithm design, and continuous model improvement. It emphasizes that the successful application of machine learning requires strategic planning, data management, and skilled personnel to realize its potential for innovation and competitive advantage.

Running head: MACHINE LEARNING FOR MERCHANDISE BUSINESS
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Name of student
Name of university
Author’s note:

1
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Executive summary
The report has investigated the application of machine learning for merchandise business for
improving the business process of JD. Due to lack of application of advanced technologies,
they lacks innovation in business, especially if compared to their competitors and therefore
they want to incorporate machine learning as a potential technology for brining innovation in
their business strategies. However, after the initial assessment, it is recognised that JD as of
now does not have that required strategies that is necessary for incorporating such advanced
technology. Along with this, they do not have required infrastructure support and their
employees do not have required skills for working with such technologies. Therefore,
recommendation on how to ensure effective application of machine learning in merchandise
business is provided with detailed strategies for implementation of this technology for the
organization.

2
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Table of Contents
Introduction:...............................................................................................................................3
Application of machine learning:...............................................................................................4
1. Defining machine learning................................................................................................4
1.1. Explanation of machine learning................................................................................4
1.2. The difference and relationship between artificial intelligence and machine learning
............................................................................................................................................4
2. Analysis of impact of machine learning in different industries........................................4
2.1. Description of the application of machine learning in three different industries........4
3. Application of machine learning in JD..............................................................................5
3.1. Description of how machine learning can be adopted in JD.......................................5
3.2. Discussion of its application to different business functional areas of JD..................7
3.3. The advantages and disadvantages of its application..................................................7
4. Analysis of issues of machine learning:............................................................................8
5. Recommendations for application of machine learning in business.................................9
Conclusion:................................................................................................................................9
Recommendations:...................................................................................................................10
References:...............................................................................................................................11

3
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Introduction:
The report is designed by ABC consultancy, for JD for considering machine learning in their
organization. ABC consultancy has reputation in incorporating IT innovations in
organizations while helping clients to realise their potential and applying values to their
business through innovation.
Due to lack of application of advanced technologies, it is identified that JD lacks innovation
in business compared to their competitors and therefore they want to incorporate machine
learning as a potential technology for brining innovation in their business strategies.
The purpose of the report is to analyse the technology of machine learning for merchandise
business for enhancing the business process of JD.
The objective of the report is to define machine learning, identifying its differences with
artificial intelligence, how it is applied in different industries, application strategies
specifically designed for JD along with functional areas where machine learning could be
applied for JD. Along with this, the objective of the report is to analyse issues of machine
learning for the organization.
In order to investigate the problem, extensive research is conducted with reference to
organizational information provided by JD for preparing an investigation report for applying
machine learning in their merchandise business. Along with this, different project reports
related to machine learning in merchandise business is considered for enhancing the quality
and efficiency of the investigation of the problem.
The report first provides definition of machine learning, along with its differences with
artificial intelligence. Then it discusses application of machine learning in different
industries, application strategies specifically designed for JD along with functional areas
where machine learning could be applied for JD. The report then provides a description of
issues regarding machine learning including legal and ethical along with social issues. Along
with this, recommendation for application of machine learning is provided which is then
followed by conclusion and recommendation for this report as well.

4
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Application of machine learning:
1. Defining machine learning
1.1. Explanation of machine learning
Machine learning is considered as an important aspect of artificial intelligence where a
system is trained in such a way that it learns automatically while improving from experience
while there is no requirement to program the system explicitly (Witten et al., 2016).
Therefore, machine learning is an emerging technology in the context of computing and
artificial intelligence as well.
1.2. The difference and relationship between artificial intelligence and machine learning
Machine learning is closely related to artificial intelligence. However, there are some
significant differences between these two along with some similarities as well (Jordan &
Mitchell, 2015). Artificial intelligence while compared to machine learning, it is identified
that these technologies are considered for providing machines ability to execute tasks that are
advanced and therefore requires intelligence for execution as well.
However, in artificial intelligence machines might be required to program explicitly to
execute tasks assigned to them. However, in machines learning machines only require access
to data and they will learn themselves how to execute those tasks based on the training
provided to them
2. Analysis of impact of machine learning
2.1. Description of different applications provided by the machine learning
Three industries except online merchandise industry/ecommerce are considered in this
context and machine learning applications for these industries are described as well.
Machine learning is being applied in the automotive industry, especially for driverless cars,
where automatic control of vehicles are required. Technologies such as computer vision and
sensor fusions are being considered in driverless cars which helps them to analyse the context
of the objects and process way of control as well (Kurakin, Goodfellow & Bengio, 2016). It
is important to note that these advanced technologies are being designed through the
application of machine learning.
Another important industry where machine earning is applied extensively in the finance
sector. Identification of insight from data and preventing fraud are two important
requirements in the finance sector. In order to ensure that insight from data is retrieved

5
MACHINE LEARNING FOR MERCHANDISE BUSINESS
effectively for making efficient decisions, application of machine earning is being considered
in this context. When insight from data is retrieved efficiently, it helps in identifying
opportunities for investment, helping investors to identify when to trade for enhanced profit
and therefore it is important (Schelter et al., 2018). Data mining provides a context for
identifying clients with high-risk profiles and it also provides insight to identify fraud so that
it is possible to design strategies to mitigate financial fraud as well.
Machine learning provides approaches for data analysis and system modelling which makes it
an important technology for delivery organizations, public transportation and other
transportation organisations (Quintana et al., 2016). These organizations are deploying
machine learning for analysing data for identifying patterns and trends which is one of the
most important aspects for ensuring that the transportation business is effective and efficient
as well. Along with that transportation companies require to enhance the efficiency of routes
and identify potential issues for ensuring an enhanced profit and for this machine learning is
being considered by the transportation companies in this context.
3. Machine learning applications in JD
3.1. Description of how machine learning can be adopted in JD
In order adopt machine learning, it is required to follow proper strategies to ensure that the
application of the machine learning is effective and efficient as well.
Whether JD consider machine learning for optimizing products prices, predicting product
demands or enhancing consumer management, the process of application is similar, only the
type of data that is provided to the model for training is different and this depends on the type
of applications (Weber & Schütte, 2019). In this context, details regarding application of
machine learning for designing a price optimization model is analysed. However, as already
described, the process is similar for other applications as well.
The procedures that are required to be incorporated for the application of machine learning
include:
 Collection of data to train the machine
 Design of algorithm
 providing training to the model to optimize price
 change in the mechanism for prediction
 optimizing price for the model

6
MACHINE LEARNING FOR MERCHANDISE BUSINESS
 considering feedback loop
 incorporation of new data for the model
Whether JD consider machine learning for optimizing products prices, predicting product
demands or enhancing consumer management, the process of application is similar, only the
type of data that is provided to the model for training is different and this depends on the type
of applications.
In this context, details regarding application of machine learning for designing a price
optimization model is analysed. However, as already described, the process is similar for
other applications as well.
Machine learning for price optimization:
The strategies that are required for the machine learning application in designing the price
optimization model for the company is described in this context (Betzing, Hoang & Becker,
2018):
 Collection of data to train the machine: an pricing model needs to be pre-trained with
data related to products of choice and their respective price range and therefore
collection of these data one of the most important requirement for incorporating
machine learning in merchandise business for JD
 Design of algorithm: the features related to the products that were considered in the
training data are required to be analysed and for this an efficient algorithm is required.
this algorithm is also required to provide the precise prediction regarding the correct
price of the product
 providing training to the model to optimize price: Now the Pricing optimization
model associated with the algorithm analyses the predictions regarding the correct
price for the customer with references to the actual prices of the products
 change in the mechanism for prediction: the merchandise algorithm that is integrated
with the machine learning technology is changes and makes adjustment to the
mechanism for prediction of prices continuously according to the requirements
 optimizing price for the model: after completing the pre-training, predictions are
measured while considering a range of selling prices while considering product
features along quality attributes as well

7
MACHINE LEARNING FOR MERCHANDISE BUSINESS
 considering feedback loop: after a product is sold, that price is considered as a new
input in the feedback loop and the pricing model is trained so that model is capable of
predicting prices that are more accurate compared to the previous prices predicted by
the pricing model
 incorporation of new data for the model: in order to increase accuracy and
effectiveness of price prediction capability of the model, new data is incorporated for
raining the model
3.2. Discussion of its application to different business functional areas of JD
The report according to the requirements of JD has specifically considered three functional
areas as defined by the organization.
Sales and pricing:
Machine learning through optimization in product prices helps in defining competitive price
for increasing sales of the products (Betzing, Hoang & Becker, 2018)
Marketing:
Machine learning helps in determining demand for a product before launching it into market
and this helps in designing marketing strategy according to the demands of the products.
Therefore, it ensures that the marketing is effective and efficient as well (Betzing, Hoang &
Becker, 2018)
Consumer management:
Machine learning helps in determining purchasing behaviour of a consumer and therefore, it
is easier to offer products to the consumer according to their requirement which will enhance
consumer satisfaction, one of the most important aspects of merchandise business (Betzing,
Hoang & Becker, 2018)
3.3. The advantages and disadvantages of its application.
Advantages:
Some of the advantages of machine learning include (Greene, Hoffmann & Stark, 2019):
 it helps in optimizing for products and increasing sales which ensures enhanced profit
 it helps in analysing product demands in the market and therefore helps in identifying
the performance of the product in the market even before it is launched in the market

8
MACHINE LEARNING FOR MERCHANDISE BUSINESS
 it helps in identifying purchasing behaviour of the customer and advertise the right
product to them according to their requirement, therefore enhancing consumer
management
Disadvantages:
Some of the advantages of machine learning include (Greene, Hoffmann & Stark, 2019):
 it requires huge investment for designing infrastructure
 it is technically complex and requires experts for designing algorithms and training
models for different applications
 it requires significant financial resources for infrastructure maintenance
4. Analysis of issues of machine learning:
Ethical issues:
Some of the ethical issues of machine learning include (Greene, Hoffmann & Stark, 2019) :
 reduction in jobs due to extensive automations
 no framework for distributing wealth created by machines and therefore inequality in
distribution of money obtained through machine intelligence
 artificial intelligence of machines might be biased towards people if they are trained
in that way as it depends on the designers
Legal issues:
Some of the legal issues of machine learning include (Abel, MacGlashan & Littman, 2016):
 no legal accountability if mistakes are due to lack of proper decision making process
of the machine
 machines collect raw data and analyse them for personalization of consumers for
identifying their personal behaviour and preferences which is not legally allowed in
many countries
Social issues:
Some of the social issues of machine learning include (Conitzer et al., 2017):
 lack of job due to automation might affect social status of the people
 only people with ownership in companies applying machine learning will have
reputation in society for their wealth

9
MACHINE LEARNING FOR MERCHANDISE BUSINESS
 it will create differences in society due to differences in wealth and social status as
well
5. Recommendations for application of machine learning in business
Review of infrastructure:
While adopting machine learning for merchandise business, it is recommended that JD
should review the IT infrastructure of the company. It is important because, many a times, it
is identified company makes huge investment in adopting new and advanced technology, but
due to lack of infrastructure support, integration of new technology with business process is
not effective compared to the investment made for that technology (Brynjolfsson & Mcafee,
2017). Therefore, this is a recommended strategy for JD to ensure that adoption of machine
learning for merchandise business is effective and efficient as well.
Availability of technical skills:
In machine learning to train machines so that they are capable of executing advanced and
complex tasks that require intelligence, effective and efficient algorithm is required.
Therefore, to ensure algorithm design, training machines with those algorithm experts are
required (Thamir & Poulis, 2015). Therefore, it is recommended to ensure that required skills
are available within the company and if not it is recommended to outsource this service.
Training to the employees:
While adopting machine learning into merchandise business, it will significantly change the
business process. However, it is important that employees of the company is aware of this
process, otherwise it will affect the execution of business activities, therefore affecting the
application of machine learning for enhancing the business process as well (Thamir & Poulis,
2015).
Conclusion:
The report has analysed the application of machine learning for merchandise business for
enhancing the business process of JD. Due to lack of application of advanced technologies,
they lacks innovation in business compared to their competitors and therefore they want to
incorporate machine learning as a potential technology for brining innovation in their
business strategies.

10
MACHINE LEARNING FOR MERCHANDISE BUSINESS
However, after the initial assessment, it is identified that JD as of now does not have that
required strategies that is necessary for incorporating such advanced technology. Along with
this, they do not have required infrastructure support and their employees do not have
required skills for working with such technologies. However, this report provides in-depth
strategies for application of machine learning for merchandise business which will ensure that
the application of machine learning is not only cost-effective, but efficient as well.
The report also provides a detailed description regarding areas of application where machine
learning could be applied and what are the possible benefits as well as drawback of these
applications. Therefore, this report will provide a comprehensive idea to the management of
JD regarding the application of this machine learning in their organization, while providing a
detailed description about the benefits and issues of this application from organizational
perspective as well.
Recommendations:
In order to ensure that investment for infrastructure development is effective, it is
recommended to consider infrastructure review report provided by ABC consultancy after
completing the initial infrastructure assessment. This will ensure that investment for
infrastructure development is in accordance with the resource requirement, therefore
enhancing the efficiency of the project. Availability of technical skills as already described is
an important requirement for application of machine learning, and therefore, if required it is
recommended to consider outsourcing in this context. Along with this, it is also
recommended to involve employees in the organization throughout the project of machine
learning deployment in the organization to ensure that it is easier for the employees to work
with this technology effectively and efficiently and this is important as well.

11
MACHINE LEARNING FOR MERCHANDISE BUSINESS
References:
Abel, D., MacGlashan, J., & Littman, M. L. (2016, March). Reinforcement learning as a
framework for ethical decision making. In Workshops at the Thirtieth AAAI
Conference on Artificial Intelligence.
Betzing, J. H., Hoang, A. Q. M., & Becker, J. (2018). In-store technologies in the retail
servicescape. Proceedings of the Multikonferenz Wirtschaftsinformatik, Lüneburg,
Germany, 1671-1682.
Brynjolfsson, E., & Mcafee, A. N. D. R. E. W. (2017). The business of artificial
intelligence. Harvard Business Review.
Conitzer, V., Sinnott-Armstrong, W., Borg, J. S., Deng, Y., & Kramer, M. (2017, February).
Moral decision making frameworks for artificial intelligence. In Thirty-first aaai
conference on artificial intelligence.
Greene, D., Hoffmann, A. L., & Stark, L. (2019, January). Better, nicer, clearer, fairer: A
critical assessment of the movement for ethical artificial intelligence and machine
learning. In Proceedings of the 52nd Hawaii International Conference on System
Sciences.
Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and
prospects. Science, 349(6245), 255-260.
Kurakin, A., Goodfellow, I., & Bengio, S. (2016). Adversarial machine learning at
scale. arXiv preprint arXiv:1611.01236.
Quintana, M., Menéndez, J. M., Alvarez, F., & Lopez, J. P. (2016). Improving retail
efficiency through sensing technologies: A survey. Pattern Recognition Letters, 81, 3-
10.
Schelter, S., Biessmann, F., Januschowski, T., Salinas, D., Seufert, S., Szarvas, G., ... &
Deshpande, A. (2018). On Challenges in Machine Learning Model
Management. IEEE Data Eng. Bull., 41(4), 5-15.
Thamir, A., & Poulis, E. (2015). Business intelligence capabilities and implementation
strategies. International Journal of Global Business, 8(1), 34.

12
MACHINE LEARNING FOR MERCHANDISE BUSINESS
Weber, F., & Schütte, R. (2019). A domain-oriented analysis of the impact of machine
learning—the case of merchandiseing. Big Data and Cognitive Computing, 3(1), 11.
Witten, I. H., Frank, E., Hall, M. A., & Pal, C. J. (2016). Data Mining: Practical machine
learning tools and techniques. Morgan Kaufmann.