International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 5 Issue 6, September-October 2021 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Artificial Intelligence and Human-Computer Interaction

Matthew N. O. Sadiku1, Uwakwe C. Chukwu2, Abayomi Ajayi-Majebi3, Sarhan M. Musa1
1
Roy G. Perry College of Engineering, Prairie View A&M University, Prairie View, TX, USA
2
Department of Engineering Technology, South Carolina State University, Orangeburg, SC, USA
3
Department of Manufacturing Engineering, Central State University, Wilberforce, OH, USA

ABSTRACT How to cite this paper: Matthew N. O.

Computers are becoming ubiquitous and are playing significant roles Sadiku | Uwakwe C. Chukwu | Abayomi
in our lives. Domestic digital devices for leisure and entertainment Ajayi-Majebi | Sarhan M. Musa
are becoming increasingly important. To be usable, every computing "Artificial Intelligence and Human-
Computer Interaction" Published in
device must allow for some form of interaction with its user. The
International
human-computer interaction is the point of communication between
Journal of Trend in
the human user and the computer. AI has been gradually being Scientific Research
incorporated into human-computer interaction (HCI). As AI systems and Development
become more and more ubiquitous, it is imperative to understand (ijtsrd), ISSN:
those systems from a human perspective. This paper provides an 2456-6470,
introduction to the “marriage” between HCI and AI. Volume-5 | Issue-6, IJTSRD47491
October 2021,
KEYWORDS: artificial intelligence, human-computer interaction pp.811-818, URL:
www.ijtsrd.com/papers/ijtsrd47491.pdf

Copyright © 2021 by author (s) and

International Journal of Trend in
Scientific Research and Development
Journal. This is an
Open Access article
distributed under the
terms of the Creative Commons
Attribution License (CC BY 4.0)
(http://creativecommons.org/licenses/by/4.0)

INTRODUCTION
The earliest user interface to a computer was the between social sciences (on the human side) and
punched card devised by Joseph Marie Jacquard in computer science (on the computer side).
1804 [1]. Computers have been interacting with
Artificial intelligence (AI) is the use of computer
humans as restaurant servers, customer service science to develop machine that can be trained
representatives, health care providers, and sexual
to learn, reason, communicate, and make human-like
partners. In 2011, 15-million people watched on TV
decisions. It is a technology that is rapidly being
as an IBM computer, named Watson, beat human
adopted in many industries to improve performance,
contestants [2]. precision, time efficiency, and cost reduction.
Although people have interacted with computers from Although AI is a branch of computer science, there is
the beginning, it took a while for human–computer hardly any field which is unaffected by this
interaction (HCI) to be recognized as a field. HCI is technology. Common areas of applications include
concerned with understanding how people interact agriculture, business, law enforcement, oil and gas,
with machines in general and with computing systems banking and finance, education, transportation,
in particular. It is the study of how people design healthcare, automobiles, entertainment,
computer systems and how computers affect manufacturing, speech and text recognition, facial
individuals and society. It focuses on the interfaces analysis, and telecommunications [4]. Today, the
between human users and computing devices [3]. HCI fields of AI and HCI are converging.
was formerly known as the man-machine studies or As shown in Figure 1, the relationship between AI
man-machine interaction. It is an ever expanding and HCI falls somewhere into the spectrum between
multidisciplinary field which sits at the intersection human-computer interaction and intelligent systems

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 811
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
[5]. Although some view AI and HCI as having Expert Systems: An expert system (ES) (or
opposing views of how humans and computers should knowledge-based system) enables computers to
interact, we believe in unifying AI and HCI. The make decisions by interpreting data and selecting
underlying divide is not between AI and HCI, but between alternatives just as a human expert would
between rationalistic and design approach [6]. The do. It uses a technique known as rule-based
two fields explore the nexus of computing and inference in which rules are used to process data.
intelligent behavior. Although several factors work to
Neural Networks: These computer programs
keep the fields of AI and HCI apart, the AI and HCI
identify objects or recognize patterns after having
communities need to make AI ready for human been trained. Artificial neural networks (ANNs)
interaction. AI and HCI can mutually benefit from a
are parallel distributed systems consisting of
closer collaboration.
processing units (neurons) that calculate some
This chapter provides a brief introduction to mathematical functions. The ANN model
application of artificial intelligence on human- represents nonlinear relationships which are
computer interaction. It begins by providing a general directly learned from the data being modeled.
overview on both artificial intelligence and human- Neural networks are being explored for healthcare
computer interaction. Then, it presents the various applications in imaging and diagnoses, risk
applications of combining the two fields. It discusses analysis, lifestyle management and monitoring,
the benefits and challenges of AI and HCI. The last health information management, and virtual
section concludes with comments. health assistance.
OVERVIEW ON ARTIFICIAL INTELLIGENCE Natural Language Processors: Computer
The term “artificial intelligence” (AI) was coined in programs that translate or interpret language as it
1956 by John McCarthy during a conference held on is spoken by normal people. Natural language
this subject. AI is the branch of computer science that processors (NLP) techniques extract information
deals with designing intelligent computer systems that from unstructured data. NLP targets at extracting
mimic human intelligence. The ability of machines to useful information from the narrative text to assist
process natural language, to learn, to plan makes it decision making. NLP includes applications such
possible for new tasks to be performed by intelligent as speech recognition, text analysis, translation
systems. The main purpose of AI is to mimic the and other goals related to language. There are two
cognitive function of human beings and perform basic approaches to NLP: statistical and semantic.
activities that would typically be performed by a Healthcare is the biggest user of the NLP tools.
human being. AI is stand-alone independent NLP has been used in the clinical setting for
electronic entity that functions much like human capturing, representing, and utilizing clinical
expert. Today, AI is integrated into our daily lives in information [10].
several forms, such as personal assistants, automated Robots: Computer-based programmable
mass transportation, aviation, computer gaming, machines that have physical manipulators and
facial recognition at passport control, voice sensors. For example, medical robots can help
recognition on virtual assistants, driverless cars, with surgical operations, rehabilitation, social
companion robots, etc. [7]. interaction, assisted living, etc. Robotic-guidance
An important feature of AI technology is that is can is becoming common in spine surgery [11].
be added to existing technologies. AI has benefited Fuzzy Logic: Reasoning based on imprecise or
many areas such chemistry and medicine, where incomplete information in terms of a range of
routine diagnoses can iitiated by AI-aided computers. values rather than point estimates. Fuzzy logic
It embraces a wide range of disciplines such as deals with uncertainty in knowledge that
computer science, engineering, machine learning, simulates human reasoning in incomplete or fuzzy
chemistry, biology, physics, astronomy, neuroscience, data. The fuzzy model is robust to parameter
and social sciences. changes and tolerant to impression.
AI is not a single technology but a range of Machine Learning: Algorithms to make
computational models and algorithms. The major
predictions and interpret data and “learn”, without
disciplines in AI include expert systems, fuzzy logic,
static program instructions. ML is a statistical
and artificial neural networks (ANNs), machine
technique for fitting models to data and training
learning, deep learning, natural language processing,
models with data. Machine learning (ML) extracts
computer vision, and robotics. The various computer-
features from input data by constructing analytical
based tools or technologies that have been used to
data algorithms and examines the features to
achieve AI’s goals are the following [8,9]:

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 812
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
create predictive models. The most common ML come first. The design of HCI should be user-
algorithms are supervised learning, unsupervised centered or user-friendly. This requires careful
learning, reinforcement learning, and deep analysis of user characteristics and the goals that
learning. Machine learning is widely used in users want to achieve. Humans have emotion and
human-computer interaction [12]. emotion can influence important their behaviors
including technology use and customer loyalty.
Deep Learning: A subset of machine learning
While natural language input by the human
built on a deep hierarchy of layers, with each
appears desirable in human-computer
layer solving different pieces of a complex
communication, this assumes that the human can
problem. It aims at increasing the capacity of
express his knowledge and that the computer can
supervised and unsupervised learning algorithms
interpret the language.
for solving complex real-world problems by
adding multiple processing layers. An illustration The Computer: This refers to any computing
of deep learning with two hidden layers is in technology ranging from desktop computers, to
Figure 2 [13]. Figure 3 shows how AI large scale computer systems. Before we used to
encompasses machine learning and deep learning deal with stand-alone computers, but today
[14]. computers are embedded in several devices.
Data Mining: This deals with the discovery of Devices such as mobile phones or VCRs may also
be regarded as computers. The rapid development
hidden patterns and new knowledge from large
databases. Data mining exhibits a variety of of computers has made effective HCI essential.
algorithmic tools such as statistics, regression The Interaction: Humans interact with
models, neural networks, fuzzy sets, and computers in several ways using different
evolutionary models. interfaces such as keyboards, mice, tablets, voice,
and hand gestures. For example, desktop
Each AI tool has its own advantages. Using a
applications use handheld computers graphical
combination of these models, rather than a single
user interfaces (GUI), while speech recognition
model, is recommended. AI technologies are
drastically influencing the retail industry and systems use voice user interfaces (VUI). In a
customer experience. An emerging area of interest in world where the user interface has been mostly
AI is to make AI agents cooperate with each other. visual, focus has been on the GUI. The interaction
may occur at the physical, social, and
FUNDAMENTALS OF HUMAN-COMPUTER psychological context. Many other paradigms for
INTERACTION human-computer interaction exist. Perhaps one of
As shown in Figure 4 [15], human–computer the best-known paradigms is the World Wide
interaction (HCI) draws on psychology, social and Web with the browser and HTML
cultural anthropology, human geography,
ergonomics, artificial intelligence, and potentially any We must understand these parts in order to
understand HCI. Applications of HCI include
discipline that deals with human behaviors. While
computer science is the hard science of HCI, information kiosks, computer terminals, interactive
psychology is its soft science [16]. The term “human- Web sites, computer games, mobile devices, robots,
computer interaction” was popularized by Stuart K. computer games, NLP, eye-tracker, hand gesture,
fatigue monitoring, improving electronic medical
Card, Allen Newell, and Thomas P. Moran in their
records (EMR), natural language interfaces, human
1983 book, The Psychology of Human-Computer
Interaction. The main goal of HCI is to produce speech recognition, and surgery [17].
usable and safe computer systems. Usability (or APPLICATIONS OF AI IN HCI
usefulness) is often defined in terms of efficiency, The AI-related technologies are silently redefining the
effectiveness and satisfaction. The success of any way we make our intentions known to computers.
product is determined by the satisfaction of the The fields of AI and HCI are mutually influencing as
humans who use the product. The data that affect never before. The synergies of AI and HCI, especially
usability of a system include age, experience, and interactive machine learning, could become very
capabilities. relevant in several areas such as medical and business
applications. Some common applications of AI in
As its name suggests, human–computer interaction
human-computer interaction include computer vision,
(HCI) consists of three parts:
speech recognition, natural language processing [18]:
The Human: This refers to an individual user or
a group of users. The underlying objective of HCI Natural Language Processing (NLP): This s is
is that people using a computer system should the technology that enables computers to

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 813
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
understand human speech as it is spoken. NLP is Applying HCI skills would enable game
drastically changing the way we interact with developers to design applications that are tailored
more complex systems. Natural communication to the player’s cultural background, allowing
among humans consists primarily of a developers to transcend cultural barriers [22].
combination of speech, hand gestures, facial Combining the human-computer interaction and
expressions, eye motions, and body language. the artificial intelligence areas can promote
Gestures are not equivalent to speech, but they accessibility in games. This will lead them
complement each other. Conversational AI making them more democratic and useful for
systems like Alexa are constantly growing society.
smarter and better at language and conversation,
BENEFITS
all made possible through advances in machine
Human-computer interfaces have benefits both in
learning. By combining data from speech and terms of performance and in terms of the human's role
non-verbal signals computers may learn
in the system. The computer will do whatever it is
appropriate patterns of interactions for
commanded to do. It has no ego or emotion and never
cooperation between humans and computers [19]. feels overworked or underworked [23]. HCI makes a
Multimodal Interaction: This is part of big difference in terms of efficiency, convenience,
everyday human discourse. We speak, move, and even for life and death. The world is full of badly
gesture in an effective flow of communication. To designed products and some products look good but
improve reliability and usability, researchers are may not work well. The skills developed with HCI
now designing multimodal interfaces that help us to understand why some products are good
automatically learn and adapt to important and and others are bad. There are some disabled people
environmental parameters [20]. Artificial who cannot control parts of their bodies and real time
Intelligence (AI) is powering speech-based and human –computer interaction systems can help them
gesture-based systems, helping to fundamentally [24].
change HCI from simple, unimodal channels to The AI revolution is coming at a rapid speed,
complex, multimodal channels.
promising better performance than human
Computer Vision: With the progress in deep professionals. AI does not need to sleep, rest, food, or
learning and neural networks, we have made great need air to breathe. It can predict human interactions
inroads in the field of computer vision. AI and with computers and identify patterns far quicker than
computer vision have made eye tracking more humans. It is not just replacing jobs (the takeover of
efficient and affordable. human jobs by automation), it is creating new jobs as
Computer Emotion: The recent developments in well. We must decide whether to ride on the wave of
human-computer interaction have emphasized AI or resist the inevitable change. The combination of
user-centered rather than computer-centered AI and HCI can improve education and make homes
approach. This led to design of intelligent and smarter.
affective interfaces. Important discoveries in CHALLENGES
neuroscience and psychology have contributed to AI and HCI have both potential benefits and risks.
an interest in the scientific study of emotion. The biggest challenge product designers have is
Without the ability to process such affect or escaping the mindset of their own technical
emotion information, computers cannot be understanding and gaining a perspective of what it is
expected to communicate with humans in a like to be a user. HCI is a rapidly evolving discipline.
natural way. The ability of machines to recognize, Other challenges involved in HCI design include
interpret and respond according to human performing real evaluations, tailoring software, and
emotions is known as adaptive or affective recognizing design conflicts [25]. Several factors
computing. It is computing that relates to emotion impede precision in HCI devices due to a lack of
or other affective phenomena. It involves the standard for the quality of data obtained by such
study and development of systems and devices devices. Courses offered on HCI must be carefully
that can recognize and express human affects or put together so that they do not quickly become out of
emotions. It uses both hardware and software date.
technology to detect the affective state of a person
The potential negative effects of AI on society cannot
[21]. be ignored. Merely introducing human guidance into
Computer Games: Computer games are AI applications is not enough to prevent its
becoming popular and the encounter between unintended consequences. Human involvement in AI
games and culture has become inevitable. system design and development is critical to ensure

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 814
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
that AI-based systems work as intended. Some AI [5] “List of papers and talks.”
applications were very expensive and failed due to a https://people.ok.ubc.ca/bowenhui/pubs.html
lack of use value. It would be a security threat if AI [6] T. Winograd, “Shifting viewpoints: Artificial
could not be fully controlled by humans. Although AI
intelligence and human-computer interaction,”
is advancing rapidly, ethics and security are not [26]. Artificial Intelligence, vol. 170, 2006, pp. 1256-
CONCLUSION 1258.
Much of the research effort in HCI seeks to improve [7] Y. Mintz and R. Brodie, “Introduction to
human-computer interaction by improving the
artificial intelligence in medicine,” Minimally
usability of human-computer interfaces. It is a young
Invasive Therapy & Allied Technologies, vol.
field with considerable success. Researchers have 28, no. 2, 2019, pp. 73-81.
begun to be more concerned about the need for a
stronger “marriage” between HCI and AI. [8] R. O. Mason, “Ethical issues in artificial
intelligence,” Encyclopedia of Information
As AI systems become more and more ubiquitous, it
Systems, vol 2, 2003, pp. 239-258.
is imperative to understand those systems from a
human perspective. This will help the designers figure [9] A. N. Rames et al., “Artificial intelligence in
out how to build collaborative relationships between medicine,” Annals of the Royal College of
humans and computers to enhance human creativity. Surgeons of England, vol. 86, 2004, pp. 334–
The future HCI will definitely shape around AI. A 338.
new field of “Human Artificial Intelligence [10] M. N. O. Sadiku, Y. Zhou, and S. M. Musa,
Interaction” will emerge from this. For this reason, “Natural language processing in healthcare,”
HCI has gradually become part of the curricula in International Journal of Advanced Research in
several institutions [27]. The HCI courses should Computer Science and Software Engineering,
incorporate AI technology. Allowing AI to vol. 8, no. 5, May 2018, pp. 39-42.
revolutionize education will mean that it will take less
time for students to learn new skills after graduation. [11] “No longer science fiction, AI and robotics are
What the future holds is anyone’s guess. transforming healthcare,”
https://www.pwc.com/gx/en/industries/healthca
More information about HCI can be found in several re/publications/ai-robotics-new-
books in [28-40] and several others available on health/transforming-healthcare.html
Amazon.com. One should also consult the following
journals exclusively devoted to AI and HCI: [12] V. S. Moustakis and J. Herrmann, “Where do
Human Computer Interaction machine learning and human-computer
Computers in Human Behavior interaction meet?,” Applied Artificial
ACM Transactions on Computer-Human Intelligence, vol. 11, no. 7-8, 1997, pp. 595-
Interaction 609.
Artificial Intelligence [13] F. Jiang et al., “Artificial intelligence in
Applied Artificial Intelligence healthcare: Past, present and future,” Stroke
REFERENCES and Vascular Neurology, 2017.
[1] B. A. Nardi and J. Kallinilkos, “Human– [14] C. Shorten, “Machine learning vs. deep
computer interaction,” in W. Donsbach (ed.), learning,” September 2018,
The International Encyclopedia of https://towardsdatascience.com/machine-
Communication. John Wiley & Sons, 2009. learning-vs-deep-learning-62137a1c9842
[2] M. N. Posard, “Status processes in human- [15] B. K. Chadktaborty et al., “a review of
computer interactions: Does gender matter?” constraints on vision-based gesture recognition
Computers in Human Behavior, vol. 37, 2014, for human-computer interaction,” IET
pp. 189–195. Computer Vision, vol. 12, no. , November
[3] M. N. O. Sadiku, C. M. M. Kotteti, and S. M. 2017.
Musa,”Human-computer interaction: A [16] A. Newell and S. K. Card, “The prospects for
primer,” International Journal of Trends in psychological science in human-computer
Research and Development, vol. 5, no. 3, May- interaction,” Human Computer Interaction, vol.
June 2018, pp. 169-171. 1, 1985, pp. 209-242.
[4] M. N. O. Sadiku, "Artificial intelligence", IEEE [17] T. F. Waddell, B. Zhang, and S. S. Sundar,
Potentials, May 1989, pp. 35- 39. “Human–Computer Interaction,” in C. R.

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 815
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Berger and M. E. Roloff (eds.), The students' motivation, the case of human-
International Encyclopedia of Interpersonal computer interaction course,” Computers in
Communication. John Wiley & Sons, 2016. Human Behavior, vol. 72, 2017, pp. 692-700.
[18] B. Dickson, “How artificial intelligence is [28] A. Dix et al., Human-Computer Interaction.
revolutionizing human-computer interaction,” May Essex, UK: Pearson Education Limited, 3rd
2017 in Artificial Intelligence, Edition, 2004.
https://thenextweb.com/artificial- [29] A. Zahirovic et al. (eds.), The Encyclopedia of
intelligence/2017/05/10/artificial-intelligence- Human Computer Interaction: The Interaction
revolutionizing-human-computer-interaction/ Design Foundation, 2nd edition, 2013.
[19] M. N. O. Sadiku, Y. Zhou, and S. M. Musa, [30] S. Smith-Atakan, Human-Computer
“Natural language processing,” International Interaction. Thomson, 2006.
Journal of Advances in Scientific Research and [31] A. Sears and J. A. Jacko, The Human-Computer
Engineering, vol. 4, no. 5, May 2018, pp. 68- Interaction Handbook: Fundamentals, Evolving
70. Technologies and Emerging Applications. Boca
[20] Z. Obrenovic and D. Starcevic, “Modeling Raton, FL: CRC Press, 2nd ed., 2007.
multimodal human-computer interaction,” [32] Y. Rogers, H, Sharp, and J. Preece, Interaction
Computer, Sept. 2004, pp. 65-72. Design: Beyond Human-Computer
[21] M. N. O. Sadiku, A. E. Shadare, and S. M. Inrteraction. Chichester, UK: John Wiley &
Musa,” Affective computing,” International Sons, 2011.
Journal of Trend in Research and [33] D. A. Norman, and S. W. Draper (eds.), New
Development, vol. 5, no. 6, Nov. -Dec. 2018, Perspectives on Human-computer Interaction.
pp. 144-145. Boca Raton, FL: CRC Press, 1986.
[22] J. Chakraborty et al., “The human–computer [34] S. K. Card, T. P. Moran, and A. Newell, The
interaction of cross-cultural gaming Strategy,” Psychology of Human-Computer Interaction.
Journal of Educational Technology Systems, Boca Raton, FL: CRC Press, 2008.
vol. 43, no. 4, 2015, pp. 371–388.
[35] B. A. Nardi, Activity Theory and Human-
[23] W. B. Rouse, “Human-computer interaction in Computer Interaction. Cambridge, MA: MIT
multitask situations,” IEEE Transactions on Press, 1996.
Systems, Man, and Cybernetics, Vol. 9L1
SMC-7, no. 5, May 1977, pp. 384392. [36] J. M. Carroll, Human-Computer Interaction in
the New Millennium. New York, NY: ACM
[24] D. Aksu and M. A. Aydın, “Human computer Press/Addison Wesley Publishing Co., 2001.
interaction by eye blinking on real time,”
Proceedings of the 9th International [37] S. Love, Understanding Mobile Human–
Conference on Computational Intelligence and Computer Interaction. Elsevier Butterworth-
Communication Networks, 2017, pp. 135-138. Heinemann, 2005.

[25] G. Fisher, “Human-computer interaction [38] J. Jacko (ed.), Human-Computer Interaction

software: Lessons learned, challenges ahead,” Handbook. Boca Raton, FL: CPC Press, 3rd ed.,
IEEE Software, vol. 6, no. 1, 1989, pp. 44-52. 2012.

[26] A. Miller, “The intrinsically linked future [39] G. J. Kim, Human-Computer Interaction:
for human and artificial intelligence Fundamentals and Practice. Boca Raton, FL:
interaction,” Journal of Big Data, vol 6, 2019. CRC Press, 2015.

[27] J. Urquiza-Fuentes and M. Paredes-Velasco, [40] A. Sarmento (ed.), Issues of Human Computer
“Investigating the effect of realistic projects on Interaction. IRM Press, 2005.

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 816
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Figure 1 The relationship between AI and human-computer interaction [5].

Figure 2 An illustration of deep learning with two hidden layers [13].

Figure 3 Artificial intelligence encompasses machine learning and deep learning [14].

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 817
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Figure 4 Human-computer interaction and related research fields [15].

@ IJTSRD | Unique Paper ID – IJTSRD47491 | Volume – 5 | Issue – 6 | Sep-Oct 2021 Page 818