6.1. Rule-Based Systems
Strictly depending on the KBSs are the Rule Based Systems also known as
production systems.
Rule-based Systems are the simplest form of artificial intelligence and are
representing knowledge in terms of a set of rules coded into a system that
tells exactly what to do in different situations.
A formal definition is provided by C. Grosan and A. Abraham: “A rule-
based system is a way of encoding a human expert’ s knowledge in a fairly
narrow area into an automated system. A rule-based system can be simply
created by using a set of assertions and a set of rules that specify how to act
on the assertion set”.
70
It is a cyclic application of rules that operates on a working memory
containing the facts in order to produce new facts from known facts.
A Rule Based-System contains a set of rules that specify both activation
conditions and they effects; in particular, each rule is represented by the if-
then production rules, by which starting from some conditions we are
obtaining actions.
Not surprisingly, C. Grosan and A. Abraham are stating that any rule-based
system consists of three simple elements that are:
“1. A set of facts,
2. A set of rules,
3. A termination criterion,
where for termination criterion it is meant the condition that determines that
a solution has been found or that none exists and that it is necessary to
C. Grosan and A. Abraham, Rule-Based Expert Systems cit p.149.
70
53
terminate some rule-based systems that find themselves in infinite loops
otherwise”.
71
7. Risks and Opportunities of Artificial Intelligence
In recent years, the field of Artificial Intelligence has been developed more
and more and has created an unprecedented debate dividing public opinion
by splitting those who are seeing the “digital revolution” and the automation
of work as not only a risk, but also a perspective of growth from those
thinking that AI will be the next “digital divide” and thus create an
72
irreconcilable split among those who have and who has no access to such
resources.
Many and very advanced are the areas of the application of AI: from the
latest generation of digital assistants, through the self driving cars till the
chatbots, intelligent agents that have the task of communicating and being
available 24 hours a day.
The latter, despite being born many years ago, have undergone a new
youthness in the last five years, since they started to be embedded into
dialogue systems, as in online assistance software, adding to these the
possibility of engaging in various conversations.
Ibidem, cit p. 150.
71
According to G. Contissa, “the term digital divide is used to identify the different
72
situations of, on the one hand, those who can use new information-and-communication
technologies, because they have access to devices and technologies and the knowledge to
use them and, on the other hand, those who do not have the access and knowledge” in
Information Technology for the Law, cit. p. 4.
54
Additionally, the most used messaging platforms, such as Telegram,
Whatsapp and Messenger, started to made possible the incorporation of
these programs; it is not by chance that in mid-2018 has been recorded that
people are using messaging apps more than social networks.
73
More in general, according to a new report conducted by Grand View
Research, the global chatbot market is expected to reach 1.23 billion dollars
by 2025, with an annual growth of the sector of 24.3%.
74
Who is well aware of the fact that in the upcoming years, both AI and
human-machine interaction will become pivotal, is Elon Musk, the
visionary entrepreneur, leader in the automotive business, founder of Space
X, co-founder of PayPal and OpenAI, that right now is setting the agenda in
this field, in such a way to be deemed at the heart of this “fourth industrial
revolution”; not for nothing he co-founded Neuralink.
The starting idea of this corporation is to study human brain diseases by
implanting a microchip inside our brain, having as an ulterior incentive to
make the brain able to acquire information and download thoughts from it.
75
Musk sees AI as a threat to humanity if left unchecked and keeps telling us
that the long-term goal is to achieve “symbiosis with AI”, in order to
prevent machines from being able to subdue humans.
76
Alex Bar, “Chatbot quanto è profonda la tana del bianconiglio?” available online in
73
Leadchampion.com, april 2018.
Sherry James, “Chatbot Market Size to Reach $1.25 Billion by 2025 | CAGR: 24.3%:
74
Grand View Research, Inc. | Markets Insider, San Francisco, September 2017, available
online in markets.businessinsider.com.
T. Lacoma, Everything you need to know about Neuralink: Elon Musk’s brainy new
75
venture, available online on digitaltrends.com, November 2017.
I. A. Hamilton, Elon Musk believes AI could turn humans into an endangered species
76
like the mountain gorilla, Business Insider, November 2018.
55
Finally, is possible to say that AI is so wide spreaded that its impact is
visible and tangible in several areas that will be analysed, in nuce, in the
next sub-chapters.
7.1. AI and Road Transport
Artificial intelligence is expected to improve mobility by enabling Smart
City initiatives to succeed and by make on-demand public transport services
a reality.
Indeed, despite the impact that AI is having into aviation by creating new
system of air traffic management, and also into railway transportation, the
area that creates more debate is the one concerning self driving cars and
public transport.
Auto-driven cars are expected to remove human errors, which are the main
cause of accidents.
According to a report of EU Commission “human error such as speeding,
distraction and drink-driving is involved in more than 90 % of accidents on
EU roads, in which more than 25,000 people lost their lives in 2017”.
77
Moreover, AI technologies can be applied also in traffic analysis by giving
to drivers information about traffic congestions and, thus, rescheduling the
route.
Transport management systems will manage transport demand based on
available infrastructure capacity and connectivity.
European Commission, “Road Safety: How is your country doing?”, pdf available online,
77
2017.
56
AI also makes truck platooning possible: in fact, as stated in a briefing of
European Parliamentary Research Service , is possible to couple several
78
Heavy Goods Vehicles (HGV) within a short distance among them by
creating a system that allow at one and the same time to accelerate or brake
according to the traffic pattern.
The leader of HGV’s platoon should be driven by a human, meanwhile in
those following the leader, the human driver can be involved only in case of
unexpected situations such roundabouts have to be faced or when a risk of
incidents are arising.
In the Automotive sector, those who stand out in the development of self-
driving cars are for instance Tesla, Waymo, V olkswagen Group and Nissan
Motor Co.
However, it must be said that the attempt is not entirely entrusted into our
days because self-driving cars are rooted even in the 1920s, when the
American company Houdina Radio Control built Linrrican Wonder, a radio-
controlled vehicle capable of performing a demonstration tour in New York,
between Broadway and the Fifth Avenue.
Finally there are plenty of concerns, such as the question of the
responsibility in the unlikely event of accident and, more tough, the ethical
guidelines that have to be followed by self driving vehicles when they will
have to make instant decisions.
In this regard Nicholas Evans, professor of philosophy at Mass Lowell
79
University, assisted by a team of technicians and philosophers, is using a
EU Parliament, Artificial intelligence in transport: Current and future developments,
78
opportunities and challenges, 2019 pdf available online
O. Goldhill, “Philosophers are building ethical algorithms to help control self-driving
79
cars” av.online on Magazine Quartz, February 2018
57
series of algorithms to recreate different scenarios of the trolley problem ,
80
thus to show how an autonomous AI-machine would behave according to
different ethical theories.
On the one hand, the utilitarian philosophers, believe that all the lives have
equal moral weight so the relative algorithm would assign the same value to
passengers as to pedestrians.
On the other hand, others believe that the driver has an extra moral value
and therefore, in some cases, the car is authorised to protect the driver.
The latter category included Mercedes-Benz chief executive, Christoph von
Hugo able to spell out that “if you know you can save at least one person, at
least save that one. Save the one in the car - justifying his decision by
additionally stating that - the car should prioritise lives it has more control
over and can keep safe after any potential accident”.
81
It can be said that the topic in question is very controversial and there are
many studies that have investigated the different currents of thought; what is
certain is that people must become increasingly aware of these risks and
opportunities so as not to find themselves displaced in the future.
The Trolley Problem is a thought experiment of ethical philosophy formulated in 1967 by
80
Philippa Ruth Foot which proposes an ethical dilemma that foresees a situation in which a
cart runs along high-speed tracks towards five bounded people, unable to move. However,
there is the possibility of pulling a lever to redirect the trolley to another track, where,
nevertheless, is laying down a person tied to the ground. Here, the question are, firstly,
whether act or not to act by pulling the leverage, and what is morally acceptable in between
killing a person in order to save five or letting the five people die.
E. Pollock, Machine Learning: How Should Self-Driving Cars Decide Who to Save? av.
81
online in Magazine Engeneering.com, August 2018.
58
7.2. AI and E-Commerce
Recently an out-and-out discipline has spread, namely the Artificial
Intelligence Marketing, which exploits the most modern technologies that
fall within the scope of AI, such as Machine Learning and Natural Language
Processing, integrated with mathematical and statistical techniques and
Marketing Behavioural.
In concrete terms, it is the use of the algorithms of AI and Machine
Learning with the aim of persuading people to buy a product or access a
service.
Indeed AI predictive algorithms have the main role of customising the offer,
making the entire process of supply and demand very dynamic; for instance
just think about the prices applied for airline tickets, etc.
Ictu oculi, it would seem impossible to explain how this pattern works; for
this reason it is crucial to remember the concept of Big Data (chapter 5.1).
Big Data is not just an analysis of many data.
In this specific case it is a complex process capable of extracting new
information in order to understand consumer behaviour, thereby improving
their shopping experience.
In this regard it does not appear misplaced the quotes from W.E. Deming:
“In God we trust; all others must bring data”.
82
Indeed, according to Deepak Kanakaraju, Ceo and Co-founder of
PixelTrack Digital, “intuition and gut feeling plays a major role in finding
A. Breckler, “In God we Trust, all others bring data”. Designing a Data-Informed
82
Decision Process with Edwards Deming: Grandfather of the Lean Startup, av. Online in
Medium.com, November 2013.
59
out what your customers want but intuition is helpful only when you do not
have Data”.
83
An example of how the study of data can make the difference in marketing,
is traceable back in the 1998, when Tesco (chain of British grocery stores)
carried out a study and found that customers who start buying Pampers also
start buying more beers; so they placed a six-package of beer alongside the
nappies: the beer sales arose in an exponential way.
The explanation of this “strange” behaviour, is that fathers of small children,
having no longer time to go to the pub, they were “obliged” to drink beer at
home.
84
Coming back to the present days, the most popular analyses made through
Big Data are basket analysis and cross-selling analysis.
They determine the associations between the products within a consumer
basket, increasing both the quantity of products in the same category and of
complementary or even totally different categories.
One of the tools that uses Big Data to study consumer behaviour is SAP
Hybris Marketing (Systems, Applications & Products in Data Processing)
which aims to verify what the consumer is looking for, by combining
information on what the consumer does in a given present moment and what
the consumer has done in the past.
85
SAP hybris Marketing can collect data based on what the consumer bought
in the online store, what he/she looked at on the web pages, what he/she
D. Kanakaraju, How Tesco Increased Beer Sales by Placing them Next to Diapers, av.
83
Online on linkedIn, February 2015.
BBC News, “Do diapers drive dads to drink?”, av online, April 1998.
84
Pivotree Growth Team, What is SAP Hybris Marketing? , av online on pivotree.com,
85
May 2017.
60
shared on social media, and also on requests or complaints he/she sent to the
company.
All this collected information are entered into individual customer profiles.
Moreover, by adding to all this amount of data, the virtual assistants like the
aforementioned chatbots or Siri from Apple and still Alexa from Amazon to
name a few - who exploit the AI both for the recognition of natural language
and for the analysis of user habits and behaviours - it is possible to create a
complete image of the customer's behaviour and a real-time
individualisation.
Finally, we can say that the consumer’s experience has become the new
standard.
The process of buying a product or service is no longer seen as a simple
action, but as a complex experience that can cause the consumer to return to
the company or never come back.
No wonder that big companies such Amazon are tracing every single step of
the users into the web stores without losing any information; let’s think
about the stand-by cart on a web shop, isn’t it true that, as soon as we are
not finalising a purchase, we receive an offer for a price reduction for the
same products that were in the abandoned shopping cart?
7.3. AI and Job Market
The AI can be considered as the last phase of a long automation process,
which has been underway for over 200 years.
Starting from 1760 with the use of the steam engine, the soul of the first
industrial revolution, followed in 1870 by the advent of the second
61
Ricevi informazioni sui nostri servizi, sulle offerte e non perdere news e consigli su università e lavoro.
