University of Hertfordshire: Advanced Research Paper on AI Topics

This report presents a research paper exploring three key areas within computer science: Human-Robot Interaction (HRI), Data Science, and Machine Learning. The HRI section investigates the importance of interaction between humans and robots, discussing relevant fields like AI and robotics, and the need for seamless communication. It poses research questions about the necessity of HRI, the role of machine learning, and the connection to computer science, outlining a positivism-based methodology with deductive approaches. The Data Science section examines the multidisciplinary nature of data science, its relation to computer science, and research questions regarding its importance and methodologies. It employs both qualitative and quantitative data analysis, using positivist and deductive reasoning. Finally, the Machine Learning section focuses on its applications in AI, addressing its significance, research questions, and methodologies, including positivism and inductive approaches. The paper highlights the relevance of each field in the context of modern computer science and its applications. The assignment follows the guidelines of the University of Hertfordshire's module 7COM1084, 'Advanced Research Topics in Computer Science', and includes a literature review and references.

FORMATIVE COURSEWORK 1: WRITING A
RESEARCH PAPER
MODULE CODE AND NAME: 7COM1084-O9O1-2019 AND ADVANCE RESEARCH TOPICS IN COMPUTER SCIENCE
Author: PENDYALA VASUDEVA VIVEK CHAITANYA
SRN: 17060709
XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

I. HUMAN ROBOT INTERACTION
A. Research
Area
Human-robot interaction is one of study method where
the interaction between the human and the robot. There are
several fields that is directly connected with Human Robot
interaction. Those are Artificial Intelligence, robotics,
understanding natural language, social science and design
etc. The academic speculation and science fiction can be
handled through human communication. It can be assumed
that very few tasks can be handled through robots, even
better than the human being. That is why the interaction
between human and robots are essential for completing this
futuristic process. That is why the topic will address the
specific area where the relationship between both of them are
addressed appropriately by which technologies (Lyons
2013).
B. Relation to
Computer Science in General
The computer programming procedure with the accurate
result is needed to control the behaviour of robotics and also
the advance science is required to implement the purpose.
Computer programming is directly related to computer
science which will help to combine the interaction
appropriately. Machine learning, Artificial intelligence is the
main field where the topic will be directly connected (Beer,
Fisk, and Rogers 2014).
C. Research
questions and methods
The research questions regarding Human and robot
interaction:
Q1. Why human and robot interaction is required?
Q2. How machine learning will help to make the
relationship between human and robots?
Q3. How computer science is related to the human-robot
interaction?
The methodology is used to explain the data collection
procedure to complete the research. The research will be
done with the appropriate way of the task completion, which
should be needed to specify the steps of the method. The
trial techniques will help to manage the research philosophy.
The researcher has been chosen the positive impact, which
is needed to make the sequence in better aspects
(Guadarrama et al. 2013).
There is there a kind of philosophy that should be used
to analyse the process. Those are Positivism, realism and
doxology. All of the methods will help to manage the
philosophy in scientific ways. The researcher has been
chosen the positivism to manage the social approach
appropriately. It will help to complete the data with the
scientific way, which will help to manage the situation. The
independent concept with the human mind interaction will
help to manage the situation with the better aspects.
Positivism will always help to analyze the quantitative and
qualitative approach both. The investigation with the help of
quantitative data will help to manage the situation
appropriately.
There are several processes of approaches that may fit
into the research. The researcher has been chosen the
deductive approach, which may help to manage the test
theory which can be easily done by the procedure. The
deductive approaches will always help to manage the high
structural presentation which can be done approximately the
report of the task.
D. General
relevance
Artificial intelligence is one of the main facts, which is
directly related to Human-robot interaction. In today’s
world, the programmers are trying to input codes to make
the system better with the help of the robotics. In fact, it can
be said that some field’s task can be done by robots. And the
robots may give a better result than the human being. That is
why human and robot interaction should be flawless. The
communication between them can be done in various ways
(Sheridan 2016). The communication and the interaction are
both category that can help to change the situation
appropriately. The HRI field should be managed
appropriately to make the situation better. The human and
robot interaction is hugely needed to modify the task in such
a way that the task should be maintained appropriately. The
general categories of the interaction are needed to modify
the situation. The social interaction and the cognitive
aspects are properly managed through this process. The
spaceship and the support of the task should be completed
according to task completion. The limit edge focal point
solution can be done by this deductive approach solution.
The high structural process can be solved by deductive
approaches (Tsarouchi, Makris, and Chryssolouris 2016).
The research design will help to verify the steps with a
logical and intelligible way. The blueprint of the study can
also be made if required. The issues of the research are
necessary to make the vision of the task. So, the proper
observation will help to manage the research with primary
data. The researcher has been chosen the descriptive design
for collecting the in-depth data (Lasota, Rossano and Shah
2014). The primary and secondary both data collection has
been used. The quantitative data has been taken for the
numeric data entry.
II. DATA SCIENCE
A. Research
Area
Data science is a multidisciplinary method where the
data can be fetched through Scientific methods, algorithms,
processes. Structured and unstructured both data can be
analysed by this procedure. The main data mining and the
big data are related to the data science. The main machine
learning and the data analysis with the statistics are the
pillars of the data science. The research will address the
relation between computer science, information science,
mathematical and statistics with the data science which will
help to analyse the data as per the data handling requirement
(Swan 2013).

B. Relation to
Computer Science in General
The data science is purely dependent with mathematical
support and machine learning. Both of them are the core
formation of computer science.
C. Research
questions and methods
There are some research questions are needed according
to the requirement:
Q1. Why is data science needed in today’s world?
Q2. How the machine learning and statistics will help to
manage data science?
Q3. How the data science methods can handle the
structured and instructed both data?
For the present research, both qualitative and
quantitative data are chosen. The assimilation of both types
of data is useful to develop the research through assuring
that barriers of one is balanced by other’s strength. Further,
one would assure that the understanding can be enhanced
through integrating various ways of understanding (Waller
and Fawcett 2013)
Maximum of the evaluations would collect both the
qualitative and quantitative data. Nonetheless, it is vital to
make a plan from advance regarding how those are to be
combined. It is helpful in enriching or utilizing the
qualitative tasks to determine the challenges or retrieve data
on the variables that are never gained through quantitative
works. Then there is the examining that includes the
generating of hypothesis from the qualitative tasks to be
evaluated with quantitative approaches (Kalidindi and De
Graef 2015). Next, there is explaining that denotes the
utilization of qualitative data for understanding the
unanticipated outcomes from the qualitative data. Moreover,
there is a triangulation that involves rejection or reinforcing
or confirming. It deals with the rejection or verification of
outcomes from quantitative data through qualitative data or
vice versa. Here, the positivist approach is to be taken. It
views the society to shape the individual and expect that the
social facts are able to provide shape to the individual
activities. Here, the sociologists tend to look for the
relationships and the correlations between two or more
number of variables. Further, deductive reasoning is to be
applied here (Provost and Fawcett 2013). It is the top-down
method drilling awards from the common to specific. The
approach is helpful for the market researchers to think about
the research already conducted and designed the concept
regarding amending or extending the theoretical basis. It
begins with some Axios the simple actual statements
regarding how he surrounding works. This understanding of
the phenomena could be deduced from the kind of axioms.
The deductive method includes the testing of any pre-
determined theory, hypothesis and explanation. It also helps
the researchers to assure the hypothesis by utilizing current
theories. The information existing is been dissected for
accepting or rejecting the hypothesis for achieving research
objectives (Shoro and Soomro 2015).
D. General
relevance
Data research aims to create the algorithms and systems
for retrieving knowledge, seeking patterns, create
predictions and insights from various data related to a
different application and the visualization. However, the
challenges involve the facts that terabytes to petabytes of
information comes out every day. Again, maximum of the
discipline is witnessing problems of big data analysis
including government, civil engineering, law school, bio-
informatics, life science and medical sciences. The
information is been coming from varies formats like images,
audio or video, structured data and free texts. The analysis
activities done on the data are turning to be more developed
and the high-performance platforms of computing are
advancing quickly like mobile-computing, multi-core
machines and cloud computing. The feedback and
communication requirements to get established between
people, algorithms and machines. As the present word is
becoming rising digital space, the organisations have been
dealing with the yottabytes and petabytes of unstructured
and structured daily. Above all, the evolving innovations
have made the cost savings and smarter storage spaces to
store critical data (Van Der Aalst 2016).
III. MACHINE LEARNING
A. Research
Area
Machine learning is one of the applications which is
used in Artificial Intelligence. Machine learning is mainly
focused on computer programming development. That is
how the improvement will be done automatically with the
appropriate process. The inference and the pattern of the
specific task should be modified with machine learning.
That is how the explicit task should be performed through
this procedure. The research will be focusing on the specific
field, which will specify the need for machine learning in
the modern era of computer science and usage of this
process (Jordan and Mitchell 2015).
B. Relation to
Computer Science in General
The difficult portion of the algorithm analysis in a
specific field like computer version and email filtering can
be done by machine programming. Also, mathematical
optimization and computational statistics will be analysed
through this procedure. So, it is related with computer
science is a specific field.
C. Research
questions and methods
There are specific research questions are situated
according to the requirement:
Q1. In which area machine learning is essential?
Q2. How email filtering process can be done by
machine learning?
Q3. How Artificial Intelligence will be benefitted
by machine learning?
The ordered method of data collecting for the
specific field is known as the research methodology. The
numerical value, trial techniques and the hypothetical issues
are analyzed in this portion to get the correct result of the
topic. Different instruments can be used to get the result
appropriately. The researcher has been chosen the

positivism and deductive approach to complete the research
according to the case study (Lison 2015).
The research philosophy is the main fact which will
address the data analysis and gathering process
appropriately. The main two types of data collecting process
can be followed, like epistemology and doxology. There are
mainly three types of philosophy are used, they are
Positivism, realism and interpretivism. The researcher has
been chosen the positivism. The researcher has been chosen
the positivism, which will help to make the hypothesis about
the research easily. The quantitative and qualitative of both
data analysis will be done by the positivist method. It is one
of the most essential structure processes where vital data
collection is hugely easy and appropriate (Witten et al.
2016).
The data collection steps can be calculated through
the research approach. The deductive and the inductive
research approach can be accepted to make the calculation
better. The deductive approach will always follow the
theoretical data and the inductive approach will analyse the
new theory which is recently collected. The research has
been chosen the inductive approach to complete the
research. The inductive approach has been chosen to fetch
solid proof for the justification.
The research design can be divided into three parts, those
are explanatory, descriptive and the exploratory. The
exploratory research method has been chosen by the
researcher because the issues of the research can be
highlighted to complete the situation appropriately
(Obermeyer and Emanuel 2016).
The data collection will be done by two processes, with
primary data collection and secondary data collection. The
raw or primary data can be collected through the journal,
website and articles. In the other hand, longitudinal study
will help to gather the data by the secondary method.
The data analysis can be done with the two types of data
analysis process, those are qualitative and quantitative
(Abadi et al. 2016). The quantitative data analysis will help
to predict the data appropriately to manage the research. The
random data is used to check this procedure. The researcher
has been chosen this method to analyse the fact according to
the usage of the future.
D. General
relevance
Machine learning will always be helpful for supervised
learning. The input-output specification can be easily done
by this procedure using the mapping. Also, the training data
can be easily analysed to make the analysis better. Semi-
supervised data can be managed by taking care of the
situation with the help of the task completion. The algorithm
used by the regression is actually the formation of
supervised learning. The neural network problems can be
also solved by this system. The email filtering process can
be hugely managed by the machine learning process. In the
other hand, artificial intelligence is hugely dependent on
task modification according to the completion of the task. It
can be said that the AI has a huge impact on computing
process which can be specified by the machine learning. The
decision taking capability may be wrong throughout the task
completion, which will help to complete the task
appropriately. It can be stated that machine learning is
essential for the future AI application, so it has to be used
carefully according to the requirement (Marsland 2014).
IV. CLOUD COMPUTING
A. Research
Area
Cloud computing is the advanced procedure of the data
storing process which is used by various organisation and
personal purpose also. It can be said that a huge amount of
data has been generated in various websites, social media
and every technological organisation. This huge amount of
data should be stored in such a way, that data will be
accessed fast and appropriate way according to the
requirement. That is how cloud computing is essential. The
research will help to analyse the importance of cloud
computing regarding data storage problem-solving. Also,
vital information will be also analysed to make the
description better (Hashizume et al. 2013).
B. Relation to
Computer Science in General
The cloud computing process can be only done by
specific software, which can be only made with the r
programming or other languages which is directly connected
with computer science.
C. Research
questions and methods
Several questions will address the research:
Q1. Why cloud computing is necessary for data storing?
Q2. Which technologies are used to access the cloud
computing?
Q3. How clustering will help cloud computing to save
the data?
The research methodology is the process where the data
gathering procedure for the research can be easily done with
the help of the various steps. The main data porting will be
done by the main methods of the data collecting procedure.
The researcher has been used the positivism to complete the
research (Sadiku, Musa and Momoh 2014).
There are two types of research philosophy can be
followed by the researcher. Those are doxology and
epistemology. The research philosophy has three methods
which can be selected for the research, those are Positivism,
realism and interpretivism. The positivism is used for
scientific data collection, on the other hand, interpretivism is
used for the social data collection. Realism is used for the
human mind though with the scientific way. The researcher
has been chosen the positivism to take the specific data
according to the task completion. The quantitative and
qualitative study, both can be done by the positivism that is
why the researcher has been chosen (Dinh et al. 2013).
There is two types of research can be chosen to complete
the situation appropriately. The inductive and the deductive
approaches are the processes which should be chosen. The
deductive approach should be done with the process of the
test theories which can be needed. The specific level of the
research with the scientific investigation has been done by
the researcher to get the result. It can be stated that the
researcher with this approach can get limit aged focal
solution.

The researcher should do the research design with the
help of the logical data of topics. The observation and
explanation will be included to make the task done. Issue of
the research and the cause-effect situation can be better
explained in the designing. The descriptive research design
has been chosen to make the situation better with the help of
task completion. Also, the multi-faced solution can be added
to complete this process (Rittinghouse and Ransome 2017).
The researcher has been chosen as the primary and
secondary data collection to make the research complete.
Both data collection is essential and used in this case. The
dynamic data has been collected through the primary data
collection procedure. The secondary data is used to collect
quality or performance data.
The quantitative data analysis has been chosen, which
will be used to collect the numeric data. The statistical data
of the finest approach is the fact of this kind of data
collection. The data has been chosen for the target amount
of task completion which will help to manage the situation
in better condition. The interpretive philosophy will help to
manage the situation in better conditions.
D. General
relevance
Cloud computing is one of the metaphors if internet
technology. The data programs should be managed with
appropriate way. This will help to manage the data
appropriately. First of all, a huge amount of data is stored in
the cloud for future use and access. The data should be
divided into the master and slave functionality, which will
help to fetch the appropriate data within the right time. In
the other hand, the data should be replicated into several
clusters. Cluster is one of the most essential units which will
store unique data. That is how the data should be managed
to make the situation better with the help of the study
completion (Carlin and Curran 2013). So cloud computing
is necessary for the data saving and modification of the data
process. The Apache Hadoop, Spark are the main
application process which can be modified with the help of
the task completion. The research will help to address the
importance of cloud computing in modern days. Every
situation can be handled to manage the situation with a
better perspective and advantages. Cloud computing is
modifying each and every day, that the data should be
managed easily. There are several scopes where the data can
be accessed, also the data crash will be a problem for cloud
computing. This problem should be avoided (Arora,
Parashar and Transforming 2013).
REFERENCES
[1] Beer, J.M., Fisk, A.D. and Rogers, W.A., 2014. Toward a framework
for levels of robot autonomy in human-robot interaction. Journal of
human-robot interaction, 3(2), pp.74-99.
[2] Guadarrama, S., Riano, L., Golland, D., Go, D., Jia, Y., Klein, D.,
Abbeel, P. and Darrell, T., 2013, November. Grounding spatial
relations for human-robot interaction. In 2013 IEEE/RSJ International
Conference on Intelligent Robots and Systems (pp. 1640-1647). IEEE.
[3] Lasota, P.A., Rossano, G.F. and Shah, J.A., 2014, August. Toward
safe close-proximity human-robot interaction with standard industrial
robots. In 2014 IEEE International Conference on Automation
Science and Engineering (CASE)(pp. 339-344). IEEE.
[4] Lyons, J.B., 2013, March. Being transparent about transparency: A
model for human-robot interaction. In 2013 AAAI Spring Symposium
Series.
[5] Sheridan, T.B., 2016. Human–robot interaction: status and
challenges. Human factors, 58(4), pp.525-532.
[6] Tsarouchi, P., Makris, S. and Chryssolouris, G., 2016. Human–robot
interaction review and challenges on task planning and
programming. International Journal of Computer Integrated
Manufacturing, 29(8), pp.916-931.
[7] Kalidindi, S.R. and De Graef, M., 2015. Materials data science:
current status and future outlook. Annual Review of Materials
Research, 45, pp.171-193.
[8] Provost, F. and Fawcett, T., 2013. Data science and its relationship to
big data and data-driven decision making. Big data, 1(1), pp.51-59.
[9] Shoro, A.G. and Soomro, T.R., 2015. Big data analysis: Apache spark
perspective. Global Journal of Computer Science and Technology.
[10] Swan, M., 2013. The quantified self: Fundamental disruption in big
data science and biological discovery. Big data, 1(2), pp.85-99.
[11] Van Der Aalst, W., 2016. Data science in action. In Process Mining
(pp. 3-23). Springer, Berlin, Heidelberg.
[12] Waller, M.A. and Fawcett, S.E., 2013. Data science, predictive
analytics, and big data: a revolution that will transform supply chain
design and management. Journal of Business Logistics, 34(2), pp.77-
84.
[13] Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin,
M., Ghemawat, S., Irving, G., Isard, M. and Kudlur, M., 2016.
Tensorflow: A system for large-scale machine learning. In 12th
{USENIX} Symposium on Operating Systems Design and
Implementation ({OSDI} 16) (pp. 265-283).
[14] Jordan, M.I. and Mitchell, T.M., 2015. Machine learning: Trends,
perspectives, and prospects. Science, 349(6245), pp.255-260.
[15] Lison, P., 2015. An introduction to machine learning. Language
Technology Group (LTG), 1, 35.
[16] Marsland, S., 2014. Machine learning: an algorithmic perspective.
Chapman and Hall/CRC.
[17] Obermeyer, Z. and Emanuel, E.J., 2016. Predicting the future—big
data, machine learning, and clinical medicine. The New England
journal of medicine, 375(13), p.1216.
[18] Witten, I.H., Frank, E., Hall, M.A. and Pal, C.J., 2016. Data Mining:
Practical machine learning tools and techniques. Morgan Kaufmann.
[19] Arora, R., Parashar, A. and Transforming, C.C.I., 2013. Secure user
data in cloud computing using encryption algorithms. International
journal of engineering research and applications, 3(4), pp.1922-1926.
[20] Carlin, S. and Curran, K., 2013. Cloud computing security. In
Pervasive and Ubiquitous Technology Innovations for Ambient
Intelligence Environments (pp. 12-17). IGI Global.
[21] Rittinghouse, J.W. and Ransome, J.F., 2017. Cloud computing:
implementation, management, and security. CRC press.
[22] Dinh, H.T., Lee, C., Niyato, D. and Wang, P., 2013. A survey of
mobile cloud computing: architecture, applications, and approaches.
Wireless communications and mobile computing, 13(18), pp.1587-
1611.
[23] Hashizume, K., Rosado, D.G., Fernández-Medina, E. and Fernandez,
E.B., 2013. An analysis of security issues for cloud computing.
Journal of internet services and applications, 4(1), p.5.
[24] Sadiku, M.N., Musa, S.M. and Momoh, O.D., 2014. Cloud
computing: opportunities and challenges. IEEE potentials, 33(1),
pp.34-36.