Machine Learning and Artificial Intelligence in Online Retail Industry

This assignment discusses the implementation of Machine Learning (ML) and Artificial Intelligence (AI) in the online retail industry, specifically for the Australian company JD. The report provides effective measures for JD to adopt the utilization of ML and AI in their online retail website. It also discusses the advantages and disadvantages of chatbots and virtual assistants, and the ethical, social and legal issues associated with ML on online retailer platforms.

PROFESSIONAL SKILL IN INFORMATION AND COMMUNICATION
STUDENT DETAILS

INFORMATION AND COMMUNICATION 1
Executive Summary
In this assignment, a discussion will be made for the online retailer company, JD. The
company wants to install the application of Machine Learning (ML) on its website to enhance
the customer experience and to save the time of their clients. Thus, the company approaches
to an IT consultant of a consulting company. The Chief Technology Officer of JD company
believes that the intervention of the ML will transform the future of the online retail industry.
Currently, the application of ML and Artificial Intelligence (AI) is systematically used in the
Health care industry, transportation industry and Finance sector. The purpose of this report is
to provide effective measures to the JD related to the implementation of machine learning and
artificial intelligence norms in their online retail website. Few recommendations are also
marked in this report such as JD should setup the team of quality analyst for reviewing the
working of machine learning in the website.

INFORMATION AND COMMUNICATION 2
Table of Contents
Executive Summary...................................................................................................................1
Introduction................................................................................................................................3
Definition of machine learning..................................................................................................4
Difference between machine learning and artificial intelligence...........................................4
Relationship between machine learning and AI.....................................................................5
Adoption of machine learning in different industries................................................................6
Utilization of Machine learning in the JD..................................................................................8
Visual Search.........................................................................................................................9
ML will help in predicting customer behaviour.....................................................................9
ML and AI can emerge effective chatbots and virtual assistants.............................................10
Advantages of chatbots and virtual assistants......................................................................10
Disadvantages of chatbots and virtual assistants.................................................................11
Ethical, social and legal issues assign with the ML on online retailer platforms....................12
The concept of privacy.........................................................................................................12
The social concern with ML and AI process.......................................................................12
Legal issues..........................................................................................................................12
Conclusion................................................................................................................................13
Recommendations....................................................................................................................14
References................................................................................................................................15

INFORMATION AND COMMUNICATION 3

INFORMATION AND COMMUNICATION 4
Introduction
Machine learning defines system which is able to perform tasks normally that requires
human intelligence. Machine learning effectively utilize the binary language because it is the
language which contained just two characters: 0 and 1 (Tryolabs, 2019). Software engineers
normally utilize progressively English-like dialects which are also called as elevated level
dialects. For example: Basic, C, Java, etc. to compose programs which are then converted
into machine language by a constructing agent, compiler, or mediator. JD is an Australian
company which incorporates the online retailer's services. The company sells a wide range of
merchandise, consumer electronics, books, etc. to their worldwide targeted audience.
Recently, the management of the company notices a drastic increment in their sales. The
targeted audience nearby spent A$200 million on retail goods over the JD’s online store in
the 2018-19 fiscal year. This sales ratio is 20% down from the A$250 million of the previous
year. In this assignment, there will be a discussion about the incorporation of machine
language in the retail store. Methods will be shown in this study through which the company
can adopt the utilization of machine language. Apart from this, the legal and ethical issue for
the machine language will also be discussed in this assignment. The aim of this report is to
ensure efficient and effective norms of machine learning in the online retail website of the
JD.

INFORMATION AND COMMUNICATION 5
Definition of machine learning
Machine learning is the logical investigation of calculations and factual models that
computer intelligence utilizes to perform out a particular undertaking without utilizing
express directions, depending on patterns and intervention models. It is viewed as a subset of
artificial intelligence (AI). Machine learning manufacture a numerical model dependent on
test information, known as "preparing information", to settle on forecasts or choices without
being unequivocally customized to play out the task. Machine learning calculations are
utilized in a wide assortment of utilizations. For example, email sifting and PC vision (Sra,
Nowozin & Wright, 2012).
Difference between machine learning and artificial intelligence
Artificial Intelligence is defined as something which is form by human or non-
characteristic thing and AI can comprehend or think. There is a misguided judgment that AI
is a system but it is not a system. There can be such huge numbers of the meaning of AI, one
definition can be "It is the investigation of how to prepare the PCs, so PCs can do things
which at present human can improve” (Nickel, Murphy, Tresp & Gabrilovich, 2015).
Machine learning is the learning wherein machine can learn by its own without being
expressly modified. It is a use of AI that gives framework through this the machine learning
can consequently take in and improve for a fact.
Machine learning Artificial intelligence (AI)
It is defined as the machine learning through
system’s own experience or the acquisition of
knowledge (Meng, et. al., 2016).
This is the system-based framework which
allows the machine to incorporated its own
decision.
This aim at increasing the accuracy, not the
chances of success.
The prime objective of AI is to enhance the
chances of success, not accuracy (Marsland,

INFORMATION AND COMMUNICATION 6
2014).
It can work in a basic computer system which
understands the binary language such as 0 and
1.
It can only work in smart computer systems.
Relationship between machine learning and AI
This kind of Machine Learning calculations permits programming specialists and
machines to naturally decide the perfect conduct inside a particular setting, to boost its
exhibition. Apart from this, it is connected with the AI because it is characterized by
portraying a learning issue and not by describing learning strategies. Machine learning and AI
accepted as a product operator. For example, a robot or a PC program or a bot which
interface with a powerful situation to accomplish an unmistakable objective. This procedure
chooses the activity that would give expected yield productively and quickly (Jordon &
Mitchell, 2015).

INFORMATION AND COMMUNICATION 7
Adoption of machine learning in different industries
 Transportation industry:
ML calculations are generally developing in the transportation sector to foresee,
screen, deal with the traffic, the use of these advancements drawn the enthusiasm of
numerous organizations into the automation technology. Self-Driving Cars are an
essential perspective to examine. In spite of the fact that it is still in the testing stage,
Self-driving vehicles are the fate of transportation. The calculations for such complex
item are being created for a framework to embrace new highlights like investigating
and streamlining the information gathered from an assortment of sources, steering,
mapping and exploring to the circumstances around them in reality (Jha & Topol,
2016).
 Healthcare:
The machine-learning innovation is quickening Healthcare industry into the following
level with cutting edge handling and understanding life science. ML and AI-driven
diagnostics gathers and tolerant information to analyze and recommend potential
approaches to treat sickness with precision. Computer-based intelligence does not
intend to raise new strategies in medicine, it just extends away with existing
prescription in the correct manner, it can identify restorative issue a lot quicker than a
human sense (Ghahramani, 2015).
Philips, a main tech organization in the health care industry, venturing towards the
stage to join AI and different advancements called Adaptive Intelligence, to reach and
to comprehend the necessities of medicinal services. Organizations like Babylon
Health, Infravision, Freenome, and numerous others are aiming at systematic models
to anticipate and to analyze the conditions inside the ideal opportunity for viable
treatment (Faggella, 2019).

INFORMATION AND COMMUNICATION 8
 Finance:
This industry is a pioneer of AI and ML innovation. Finance sector incorporates a
tremendous measure of valuable information, and AI calculations are
comprehensively used to play out the best forecast about bad debts, overseeing of
fraud transactions, exchanges to speculations transactions, etc. Man-made intelligence
gives vital activities for business improvement.
Man-made intelligence controls the monetary administrations and arrangements
guaranteeing a safe business. It expects to convey the administrations that a client
needs and recognizes the bank cheats. Banking segments depend intensely on these
innovations because these coordinated are utilized to advance Chatbot's and
Conversational Interface to serve the client with the pertinent data according to
prerequisites (Coussement & De Bock, 2013).

INFORMATION AND COMMUNICATION 9
Utilization of Machine learning in the JD
Clients enjoy perusing on the web indexes of online retail websites because of online
shopping websites appealingly present items and give a lot of data about them. AI
advancements can fundamentally improve clients experience to expand commitment and
effect transformation rates. By introducing the MI on the website, the MI will itself organize
the items as per the preference of the customers. For example: when a particular client will
open the website for the first time then MI intelligence will record the items for which the
client has looked. Through this, the MI will be able to organize the products as per the
matching perspective of the client. MI will react with their intelligence when the customer
will log in with their same old id and password (Carrasquilla & Melko, 2017).
The mechanism of the ML will work on the website by processing the
recommendations. Many other online retail companies such as Amazon, eBay, Flipkart, etc.
utilize ML for processing recommendations to the client because of exceptionally customized
offers and improved client experience. Recommendations commonly accelerate previous
searches of the client and make it simpler for clients to access substance they're keen on and
ML intelligence surprises them with offers they would have never looked for (Bottou, Curtis
& Nocedal, 2018).
The client begins to feel known and is bound to purchase extra items or devour
progressively content. By knowing the desires and nature of the client, needs and indicating it
immediately, it more outlandish that it leaves the stage. This converts into a higher possibility
of procurement and a diminishing in the danger of losing a client to a contender (Biamonte,
et. al., 2017).

INFORMATION AND COMMUNICATION 10
Visual Search
Clients will in general scan for visual substance before purchasing through the JP
website. Moreover, sometimes they can only with significant effort to discover great
catchphrases/keywords to depict what they need. The objective of the visual hunt is to make a
lot simpler for shoppers to discover precisely what it is they're searching for. Rather than
composing an inquiry. For example: collar white mischief t-shirt for men with buttons', which
will probably respond with a great deal of general outcomes. With the colossal and expanding
measure of snapping and sharing pictures, ML calculations can as of now accomplish
stunning outcomes (Beam, AL & Kohane, 2018).
ML will help in predicting customer behaviour
The objective of a foreseeing client conduct/behaviour system is to gauge how
purchasers will carry on, later on, dependent on information with his previous keywords. For
example: previously a client searched for “men’s t-shirt” that does not mean that client will
again search for “men’s t-shirt”. It is possible that this time client searches for "men's shirt"
or "men's clothing accessories"; so, here ML system should be capable to predict the future
searches of the clients through his/her past behaviour (Apiumhub.com, 2019).
These frameworks enable JP to divide their clients in the specific segment and
perform customized activities that are more compelling than general approaches.
Additionally, taking activities dependent on anticipated client needs expands customer’s
retention and loyalty. For instance, to know which clients are probably going to make a buy
in the following 7 days. Increasingly mind-boggling forecasts may have to do with significant
occasions in individuals' lives. Anticipating the requirements of customers is a difficult
errand where ML intelligence calculations are of incredible assistance.

INFORMATION AND COMMUNICATION 11
ML and AI can emerge effective chatbots and virtual assistants
Chatbots cooperate with clients and mimic a human discussion, carrying them closer
to the shopping that purchasers get in a physical store. They can give included an incentive at
various levels. For instance, chatbots can be utilized in the JD’s website to empower extra
customer satisfaction and to improve searching capacities over the catalogue (Tryolabs,
2019).
On the other hand, the objective of a virtual assistant is to imitate a human sales
executive who can assist a customer in finding what they are searching for. A virtual assistant
is great to enhance the overall satisfaction of the client. Using an advanced level of the
machine learning process and artificial intelligence also enhances the market capacity and
reputation of the company among the other competitors because it provides the customer with
an advanced level of satisfaction (Sra, Nowozin & Wright, 2012).
Advantages of chatbots and virtual assistants
 These tools are high power with the power of ML and AI resultants in saving the time
of customer through their instant response behaviour.
 Chatbots do not convince or try to sell anything to the customer instead they only
process the answer of the customer's query.
 These assistant work for 24*7 (Nickel, Murphy, Tresp & Gabrilovich, 2015).
Chatbots
Virtual Assistants

INFORMATION AND COMMUNICATION 12
Disadvantages of chatbots and virtual assistants
 Chatbots are regularly observed to be muddled and require a ton of time to
comprehend the client's prerequisite.
 Chatbots are introduced with the rationale to save time and improve client association.
However, due to the restricted information accessibility and additional time required
for updating, this procedure shows up additional time-taking and costly.
 The development and research time for the chatbots and virtual assistants are the
costly process because every insistence of the ML needs to specifically programmed
(Meng, et. al., 2016).

INFORMATION AND COMMUNICATION 13
Ethical, social and legal issues assign with the ML on online retailer platforms
The concept of privacy
This lies in the ethical framework. People have observed that websites or mobile
application which are enabled with the advance ML system are taking the data of the people
and the companies are selling customer’s data (their name, email address, phone number and
location) to the other big companies. People had experienced that whenever they insert with
information on the retailer websites then immediately, they start receiving calls or messages
for different kind of offers or deals (Jordon & Mitchell, 2015).
The social concern with ML and AI process
In general, the employs of the online retailer platforms are concern about their
employment because people are thinking that the intervention of the AI can take away their
employment.
Legal issues
The uses of the AI give birth to the legal concern as the Federal government of
Australia made the specific cyber laws that state the personal information of the general
public should not be remotely transferred by the ML system in the website. Thus, the
advanced system needs to be developed for the JD (Jha & Topol, 2016).

INFORMATION AND COMMUNICATION 14
Conclusion
It has been concluded that the development of the ML in the online retail website of
JD will help them in providing better customer experience than their other competitors. The
AI can process intelligence recommendation to the clients and also analyse customer
behaviour. Apart from this, the ML and AI system help in saving the time of the customer
and also provides specific suggestions but this system incurred heavy setup cost. The
government is a concern with the privacy of their citizen thus they have made the specific
privacy act. The people are also concern that the introduction of the ML and AI can take
away their employment. Apart from this, it has also been concluded in this report that AI is
pushing the limits of machine-empowered functionalities. This front-line innovation
encourages machines to act with a level of self-sufficiency, bringing about compelling
execution of iterative assignments. Machine learning is the capacity of a PC program to learn
and think. Everything can be viewed as Artificial insight on the off chance that it includes a
program accomplishing something that we would ordinarily think would depend on the
knowledge of a human. The expression "human error" was conceived on the grounds that
people commit errors now and again. Computers don't commit these errors if they are
modified appropriately. With machine learning, the choices are taken from the recently
accumulated data applying certain arrangement of calculations. So, errors become less and
the shot of arriving at higher accuracy become more.

INFORMATION AND COMMUNICATION 15
Recommendations
1. JD should setup the quality team of 8-10 people who operate to monitor the
performance of the installed ML and AI system. This team should always monitor that
when customer insert the information then where the information is traveling
(Ghahramani, 2015). QA groups utilize a blend of computerized testing and manual
testing. With manual testing, a real client tests the framework subsequent to perusing
your prerequisites. The advantage of this kind of testing in QA is that it enables them
to discover minor subtleties that don't keep benchmarks and don't coordinate with the
AI product.
2. JD should also make an effective measurement to make the ML system hack-proof.
To be valuable to programmers, machine learning tools should have the option to
make a move, make something or change themselves dependent on what they
experience when sent and how they've been prepared to respond. Hackers might not
have enough information on assaults and their results to fabricate inventive or
adaptable, self-altering models.
3. Developing digital racks through the utilization of AI technology can resultant in
enhanced customer satisfaction (Faggella, 2019).

INFORMATION AND COMMUNICATION 16
References
Apiumhub.com. (2019). AI in Retail industry: How big players increase customer
experience. Retrieved from https://apiumhub.com/tech-blog-barcelona/ai-in-retail-
industry/
Beam, A.L., & Kohane, I.S. (2018). Big data and machine learning in health
care. Jama, 319(13), 1317-1318 https://doi.org/10.1001/jama.2017.18391
Biamonte, J., Wittek, P., Pancotti, N., Rebentrost, P., Wiebe, N. & Lloyd, S. (2017).
Quantum machine learning. Nature, 549(7671), 195-200.
Bottou, L., Curtis, F.E. & Nocedal, J. (2018). Optimization methods for large-scale machine
learning. Siam Review, 60(2), 223-311 https://epubs.siam.org/doi/10.1137/16M1080173
Carrasquilla, J. & Melko, R.G. (2017). Machine learning phases of matter. Nature
Physics, 13(5), 431-435.
Coussement., K. & De Bock, K.W. (2013) Customer churn prediction in the online gambling
industry: The beneficial effect of ensemble learning. Journal of Business
Research, 66(9), 1629-1636 https://doi.org/10.1016/j.jbusres.2012.12.008
Faggella, D. (2019). Artificial Intelligence in Retail – 10 Present and future Use cases.
Retrieved from https://emerj.com/ai-sector-overviews/artificial-intelligence-retail/
Ghahramani, Z. (2015). Probabilistic machine learning and artificial
intelligence. Nature, 521(7553), 452-470
Jha, S., & Topol, E.J. (2016). Adapting to artificial intelligence: radiologists and pathologists
as information specialists. Jama, 316(22), 2353-2354
https://doi:10.1001/jama.2016.17438

INFORMATION AND COMMUNICATION 17
Jordan., M.I. & Mitchell, T.M. (2015). Machine learning: Trends, perspectives, and
prospects. Science, 349(6245), 255-260
Marsland, S. (2014). Machine learning: an algorithmic perspective. London: Chapman and
Hall.
Meng, X., Bradley, J., Yavuz, B., Sparks, E., Venkataraman, S., Liu, D., Freeman, J., Tsai,
D.B., Amde, M., Owen, S. & Xin, D. (2016). Mllib: Machine learning in apache
spark. The Journal of Machine Learning Research, 17(1), 1235-1241
Nickel, M., Murphy, K., Tresp, V. & Gabrilovich, E. (2015). A review of relational machine
learning for knowledge graphs. Proceedings of the IEEE, 104(1), 11-33
https://epubs.siam.org/doi/10.1109/JPROC.2015.2483592
Sra, S., Nowozin, S. & Wright, S.J. (2012). Optimization for machine learning. Cambridge:
Mit Press.
Tryolabs.com. (2019). The guide to Machine Learning in Retail: Applications and Use cases.
Retrieved from https://tryolabs.com/resources/retail-innovations-machine-learning/