Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Machine Learning: A Way to Smart Environment

A.M. Sunusi1,2*, Mukhtar Sabiu Yahuza3, A. Bashir4
1*
Civil Engineering Unit, Department of P.P.D. and M. Yusuf Maitama Sule University, Kano, Nigeria.
2
Department of Architectural Planning Amity University Gurgaon India
3
Department of Architecture, Near East University Nicosia (TRNC), Mersin 10 Turkey
4
Department of Civil Engineering, Federal University, Dutsin-ma, Katsina, Nigeria

Abstract:- Smart cities attempt to create a management "digital cities," and "intelligent cities" in previous
system of expanding metropolitan center to enhance the incarnations (Kitchin 2014). Despite the fact that the word
economy, energy consumption and upgrade the living "smart city" has many variations, experts have found it
conditions of residents in an effort to attain sustainability beneficial to define two broad but related meanings for the
goals. Information and Communication Technology term [1].
(ICT) now plays a significant role in decision making,
policy development and the application of new The first, is the broad integration of Information and
approaches for smart environment. The main goal of this Communication Technologies (ICTs) and Internet of Things
book paper is to highlight the importance of machine (IoT) tools into the urban environment to better monitor,
learning in the development of smart green environment assess, and manage municipal assets and services (Bibri &
in a way to achieving smart green economy. This paper Krogstie 2017b). The vast amounts of data obtained through
gives a summary on the aspect of green smart these methods undoubtedly provide a better understanding of
environment (smart cities), Machine learning and AI in the city as an ecosystem, and can be used for analysis,
the development and sustenance of smart environment modeling, and prediction (Kitchin 2014).
and the role it plays for smart green economy and
The second definition of "smartness" focuses on people,
infrastructure which are pre-requisite requirement for
smart environment. In the insight of this work, we offer with ICTs serving as tools for citizen involvement and
an explanation about AI and machine learning in smart mobilization, as well as participatory governance (Mller et al.
environment as well as the potential application of 2018). It has been proposed that a city cannot be genuinely
Machine Learning in the same. smart until it effectively harnesses the potential of social
capital, entrepreneurship, and innovation, often known as the
I. INTRODUCTION information economy (Kourtit et al. 2012).

The capacity to learn from experience, or intelligent Smart Environment is a complex system, which need a
thinking, is one of the most important aspects of human mix of transparency and context awareness. Such systems'
intelligence. The question then becomes whether machines architectures must adapt to demand space and have a modular
can be programmed to think intelligently in the same way that and flexible design capable of producing suitable services at
humans do. If that's the case, what are the fundamentals of the correct moment. Contributions from leading researchers
Machine Learning, what can it do in the construction of a describe techniques and issues related to developing and
green smart environment, and how can urban living in intelligent environments. The latest research in smart
planners/designers, architects, engineers, economists, and environment preset the philosophical and computational
others go about doing so? architecture considerations, network protocols for smart
environments, intelligent sensor networks and power line
The increasing accessibility of robust sensors and control of devices, and action prediction and identification are
actuators, as well as the convergence of technology in among the topics covered, reflecting the multidisciplinary
machine learning and pervasive computing, has sparked nature in the design of smart environment.
interest in the development of smart environments.
Researchers are also realizing that smart settings can help The characteristics of smart environments are presented
with important functions like remote health monitoring and in this paper based on the requirements established for them
intervention. The aging of the population, the high cost of and the features accomplished in prototype solutions for
formal health care, and the significance that people have on various application domains. These summaries aid in
being independent in their own homes all point to the need comprehending the importance of machine learning in
for such technologies (Parisa et al., 2011). intelligent settings. Many of our cities are adopting digital
technologies. These smart environments use data and digital
Because ubiquitous computing is so different, it is technology to enhance efficiency and inhabitants' quality of
difficult for software designers to come up with a single life, from self-driving cars to smart grids to intelligent traffic
model that can fulfill all of their needs. Smart cities" is a lights.
concept that has acquired popularity across industries and has
infiltrated the domains of sustainability, urban planning, The natural capital that cities rely on, however, it is at a
engineering, and computer science, resulting in a plethora of risk of being left behind by the digital revolution. Smart
definitions (Albino et al. 2015). Cities that were regarded technologies are already being used in environmental and
"pioneers in embedding digital infrastructure and systems resource management, with increasing technological
into their urban fabric" were referred to as "wired cities," capabilities and recent calls to action for environmental

IJISRT21OCT642 www.ijisrt.com 1259

Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
technology investments. Indeed, green infrastructure and structures [2]. AI algorithms is of high interest to smart
natural asset management are barely mentioned in smart environment builders, the so-called evolutionary computing,
environment funding mechanisms and policy initiatives. The has drawn inspiration from biological evolution. These
majority of European nations and associated partner funding algorithms perform a different type of AI learning enabling
for smart-city projects has been limited to energy, transit and machines to make autonomous decisions, adapt to a changing
mobility, and ICTs. However, in India's "Smart City mission" environment and find non-obvious solutions to complex and
aims to create hundred smart cities by implementing "smart ‘wicked’ problems[3]. Machine learning is divided into two
solutions" in the areas of e-governance, waste management, categories based on the training resources: supervised and
water and energy management, and urban mobility. These unsupervised learning.
methods, on the other hand, leave out urban green space and
forest management. As a result, the purpose of this paper was  Supervised Learning:
to highlight the importance of machine learning in attaining a The AI network is trained to map output data using a
smart green environment in the digital era. mapping function using a set of inputs and targets. The
existence of a "teacher" and the input-output data for the
II. MACHINE LEARNING instruction are the most important aspects of supervised
learning. It is further divided into two categories: regression
Despite the fact that there are numerous definitions of and classification. Supervised learning includes techniques
Machine Learning, I prefer a straightforward one. For such as support vector machines, random forecasting, and
example, machine learning (ML) can be defined as a linear regression[4]–[12].
collection of techniques that can be used to build a model
from data in order to make predictions or take actions to  Unsupervised Learning:
improve a system. Machine learning is an area of AI that deals In unsupervised learning, no direction is provided;
with the design and development of algorithms for instead, an unclassified and unlabeled input dataset is
recognizing and making decisions about complicated patterns provided for the AI network to train on in order to identify
based on experimental data. Machine-based learning models hidden patterns, responses, and distributions. Unsupervised
can be analytical in nature, predicting outcomes, or learning challenges include clustering and association,
descriptive in nature, gaining information from data. The among others. Unsupervised learning techniques such as
original purpose of machine learning techniques was to auto-encoders and k-means are common examples[13].
automate the creation of data for integration into knowledge

Fig 1:- Types of machine learning with adopted algorithms.

 Big Data variety, and velocity or the "3 Vs of big data") is amazing.
Because of the expansion of mobile phones, the Internet That indicates that by 2020, the entire amount of data will
of Things (IoT), and satellite earth observation technology, to have surpassed 44 zettabytes (or 1021 bytes), or 40 times the
mention a few sources, the rate at which we generate number of stars in the observable universe.
incredibly huge datasets every day (in terms of volume,

IJISRT21OCT642 www.ijisrt.com 1260

Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig 3: City data synchronization, collection and real-time transmission

III. MACHINE LEARNING AND SMART car, etc.) that use pervasive computing, machine learning,
ENVIRONMENT artificial intelligence, cognitive sensor systems, wireless and
sensor networking to realize cyber and physical components
AI and machine learning have the potential to transform for sensing, reasoning, and controlling the environment are
the way smart environment or cities work in a variety of ways. referred to as smart environments. Autonomous agents can
However, for an AI and ML-based smart city, developing and play an essential mediating function between human users
integrating software and hardware platforms, smart and the environment in a smart environment. This is
frameworks and algorithms, theoretical planning, and especially true when high-level cognitive skills and
mathematical computing models of ICT infrastructure is computational intelligence are utilized to cope with the
critical. This special issue encourages academicians and uncertainties of the complex environment, allowing agents to
researchers to share their work that uses AI and machine respond appropriately in a range of situations. System
learning to aid in the creation of smart environment. The integration, system maintenance, geographical and functional
availability of intelligent devices via Artificial Intelligence extensibility, social networking, mobile computing, context-
(AI) and Machine Learning (ML) has been critical in the aware applications and services, human-in-the-loop modeling
advancement of the smart city idea. You can now create a and simulation, big data analysis, cloud and edge-based IoT
cyber-physical space with traffic sensors, video cameras, frameworks and environments, field experiments and test
environmental sensors, smart meters, and other devices by beds are all issues that must be addressed in the creation of a
combining sharp computing programs with human smart environment. The primary objective of this Special
intelligence. Issue is to present and debate recent developments in the field
of sensor systems, particularly in terms of technologies,
When numerous people live in the same house, multi- architectures, algorithms, and protocols for smart
class categorization is required, which is a tough and time- environments, with a focus on real-world applications. Topics
consuming operation due to the nature of the data. that are appropriate include, but are not limited to: Wireless
Furthermore, sensor data is constantly noisy and sensitive to sensor networks (WSN) in smart cities, smart city sewage,
a variety of unknown circumstances, such as missing data and water and electricity management, Smart city healthcare
defective sensors. To achieve high accuracy classification, it service monitoring, smart city education, training and social
is necessary to identify and understand spatio-temporal services, smart transportation system planning, evaluation,
correlations between sensor readings[14]. and technologies, smart home, smart building and social
Intelligent green and smart environment (e.g., smart community networks/infrastructures, big data analytics and
home, smart factory, smart city, smart building, smart machine learning cloud computing for Internet of things,
parking, smart agriculture, intelligent traffic systems, smart machine learning etc.

IJISRT21OCT642 www.ijisrt.com 1261

 Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig 1: Architecture and components

IV. POTENTIAL USES OF MACHINE LEARNING artificial intelligence and machine learning. According to the
IN SMART ENVIRONMENT research, metropolitan regions will soon be home to 50% of
the world's population, but they will also account for 60-80%
 Smart Green Economy and Infrastructure of energy consumption and 75% of carbon emissions[28].
The depletion of conventional energy resources, city Every city has a unique chance to take a step toward greening
congestion, increased chemical, physical, and biological the global economy. There are also real possibilities for
pollution, as well as global warming, are the most serious national and local authorities to decrease carbon emissions
challenges of the twenty-first century, with all of their effects and pollution, improve ecosystems, and reduce
for quality of life. Large settlements, which are characterized environmental hazards. It is feasible to promote compact,
by high densities of population and buildings, would be the somewhat densely inhabited cities with a mixed-use urban
most affected in this regard; as a result, planners will have to shape that are more resource-efficient than any other
rethink their form and functions in order to meet the needs for settlement pattern with comparable levels of economic
a sustainable urban living [15]. Intelligent green growth production by taking advantage of the opportunity. Integrated
offers a chance to rethink our economic and smart urban and regional planning and design strategies using
environment development models as a new paradigm that technology are also available to improve urban
supports economic development while decreasing transportation, building construction, and the development of
environmental degradation and protecting natural resources. urban energy, water, and getting energy from urban waste
The conventional efficiency paradigm is giving way to a systems in such a way that they reduce reliance on natural
broader concept of societal development, one that includes resources and energy consumption in the process of long-
efficiency, fairness, and environmental sustainability[16], term planning[29]–[33].
[17], [25]–[27], [18]–[24], [24]. Growth Economic efficiency
has long been viewed as a distinct goal from equality and  How Green IT can Resolve the Issue of Sustainability
environmental goals. In light of the massive social and To develop a sustainable smart city, it is necessary to
environmental changes occurring on a global scale, an employ sustainable technologies. The demand for green IT is
increasing number of cities have shifted their development heightened by the issue of sustainability. Green IT allows for
strategies to smart policies aimed at sustainable mobility, the development of hardware and software applications that
building energy upgrades, increasing energy production from use fewer resources. Green IT's product lifetime idea can help
renewable sources, improving waste management, and to tackle the problem of rising demand for computer devices.
implementing ICT infrastructures. The objective is to Manufacturing a new computer, for example, will require far
transform Smart Cities into Smart Communities that can more resources than manufacturing new RAM for an existing
improve the quality of life of its residents by providing a long- computer. The generation of hazardous elements such as lead
term opportunity for cultural, economic, and social progress and mercury will be reduced if computer equipment is
in a healthy, safe, exciting, and dynamic environment. The recycled rather than discarded. Urban regions are responsible
primary message of an early copy of an online release on the for 80% of CO2 emissions. Cities are attempting to reduce
issue of a green economy' (2011) is that urban development CO2 and other greenhouse gas emissions, preserve resources,
must be substantially modified to enable the transition to a and provide the foundation for new economic activity by
green economy through the aid of modern techniques like utilizing ICTs and optimization approaches. Despite the fact

IJISRT21OCT642 www.ijisrt.com 1262

Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
that this goal encompasses a variety of sectors, transportation V. CONCLUSION
in metropolitan areas is responsible for the majority of CO2
emissions. Better urban design and changes in human habits, The world is on its way to becoming a supremely
both of which are generated and facilitated by new computerized environment, with current era technologies like
technologies, are seen as potential solutions to this problem IoT, machine learning, and artificial intelligence guiding the
[34]. Recycling will minimize the demand for new gadgets way. As previously stated, we have discussed how these
and, as a result, energy and resource consumption will be technologies can be used to plan a smart green environment
reduced. The usage of a decentralized system can help to that is both relaxing and comfortable to live in. Collaboration
minimize CO2 emissions. A decentralized system is an inter- with digital transformation experts and software testing
connected information system that can provide search results providers is essential for the successful implementation of
without sending the request to servers around the world. new technologies and IoT devices. It is predicted that the
knowledge of such specialists would be utilized to the
Cloud computing has established itself as a technology remodeling of various cities throughout the world in the near
that enables a variety of IT services. As the number of cloud- future. Complex systems characterize smart environments,
based IT services and apps grows, data centers with which necessitate a balance of transparency and context
thousands of web servers, storage, and network devices must awareness. Such systems' architectures must respond to
be built. Cloud data centers (CDC) offer a variety of services demand space and incorporate a modular and flexible design
to end users, ranging from high-performance computing to capable of producing appropriate services at the right time.
large-scale data analytics. Because of the massive scale of
cloud data centers, which are set up in multiple geographical REFERENCES
locations to serve distributed users, they account for 25% of
total IT electricity consumption[35]. [1]. S. A. Nitoslawski, N. J. Galle, C. K. van den Bosc, and
J. W. N. Steenberg, “Smarter ecosystems for smarter
Cloud Computing provides a variety of computing cities? A review of trends, technologies, and turning
services that are accessible over the Internet and provide high points for smart urban forestry,” Sustain. Cities Soc.,
availability, increased security, and reduced total cost of vol. 51, no. August, 2019, doi:
ownership. Infrastructure as a service, Platform as a service, 10.1016/j.scs.2019.101770.
Software as a service, Network as a Service, Storage as a [2]. D. Savic, “ARTIFICIAL INTELLIGENCE,” no.
Service, Sensor as a Service, and other cloud computing August, 2019.
services exist, depending on the type of resources given via [3]. “AI and Machine Learning for Smart Cities Call for
them [34]. The Internet of Things (IoT) can improve Papers.”
efficiency, performance, and through put when smart devices [4]. Q. B. Pham et al., “Potential of Hybrid Data-Intelligence
migrate outside of the cloud infrastructure environment. Algorithms for Multi-Station Modelling of Rainfall,”
Smart cities are residential areas that make systematic efforts Water Resour. Manag., vol. 33, no. 15, pp. 5067–5087,
to observe for themselves where records and communication 2019, doi: 10.1007/s11269-019-02408-3.
technologies are being kept, to accomplish environmental [5]. G. Usman, M. H. Ahmad, R. N. Danraka, and S. I. Abba,
sustainability, urban system authority, better health, “The effect of ethanolic leaves extract of
knowledge development, and network-driven advancement. Hymenodictyon floribundun on inflammatory
Cloud computing can help with energy and resource usage biomarkers: a data-driven approach,” Bull. Natl. Res.
concerns. Large and small organizations alike may migrate Cent., vol. 45, no. 1, 2021, doi: 10.1186/s42269-021-
their on-premise applications to cloud computing, halving 00586-y.
their energy usage and carbon emissions. The world is fast [6]. S. Idris, M. A. A. Musa, S. I. Haruna, U. U. A. A. G.
transforming into a smart city. People have gained comfort Usman, and M. I. A. Abba, “Implementation of soft ‑
because to information technology and smart cities, but they computing models for prediction of flexural strength of
have also released dangerous gases that contribute to global pervious concrete hybridized with rice husk ash and
warming. Existing and developing technology must be calcium carbide waste,” Model. Earth Syst. Environ.,
adapted in a way that has the least possible environmental 2021, doi: 10.1007/s40808-021-01195-4.
effect. The objective must be to achieve balanced growth [7]. H. U. Abdullahi, A. G. Usman, and S. I. Abba,
from both a social and an environmental standpoint. With the “Modelling the Absorbance of a Bioactive Compound
aid of green IT, this aim may be realized. Universities are in HPLC Method using Artificial Neural Network and
taking steps to educate students, as well as create and Multilinear Regression Methods,” vol. 6, no. 2, pp. 362–
maintain systems that are less harmful to the environment. 371, 2020.
More applications for employing green IT in all areas of [8]. G. Usman, S. Işik, and S. I. Abba, “Hybrid data-
company may be created with more educated and skilled intelligence algorithms for the simulation of
students. thymoquinone in HPLC method development,” J. Iran.
Chem. Soc., vol. 18, no. 7, pp. 1537–1549, 2021, doi:
10.1007/s13738-020-02124-5.

IJISRT21OCT642 www.ijisrt.com 1263

Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[9]. G. USMAN, S. IŞIK, S. I. ABBA, and F. MERİÇLİ, [20]. V. Nourani, G. Elkiran, and S. I. Abba, “Wastewater
“Artificial intelligence–based models for the qualitative treatment plant performance analysis using artificial
and quantitative prediction of a phytochemical intelligence - An ensemble approach,” Water Sci.
compound using HPLC method,” Turkish J. Chem., vol. Technol., vol. 78, no. 10, pp. 2064–2076, 2018, doi:
44, no. 5, pp. 1339–1351, 2020, doi: 10.3906/kim-2003- 10.2166/wst.2018.477.
6. [21]. Q. Bao, M. S. Gaya, S. I. Abba, R. A. Abdulkadir, and
[10]. G. Usman, S. Işik, and S. I. Abba, “A Novel Multi- P. Esmaili, “Modeling of Bunus regional sewage
model Data-Driven Ensemble Technique for the treatment plant using machine learning approaches,”
Prediction of Retention Factor in HPLC Method vol. 26160, pp. 1–10, 2020, doi:
Development,” Chromatographia, vol. 83, no. 8, pp. 10.5004/dwt.2020.26160.
933–945, 2020, doi: 10.1007/s10337-020-03912-0. [22]. Q. B. Pham, S. S. Sammen, S. I. Abba, B. Mohammadi,
[11]. G. Usman, S. Işik, and S. I. Abba, “A Novel Multi- S. Shahid, and R. A. Abdulkadir, “A new hybrid model
model Data-Driven Ensemble Technique for the based on relevance vector machine with flower
Prediction of Retention Factor in HPLC Method pollination algorithm for phycocyanin pigment
Development,” Chromatographia, 2020, doi: concentration estimation,” Environ. Sci. Pollut. Res.,
10.1007/s10337-020-03912-0. vol. 28, no. 25, pp. 32564–32579, 2021, doi:
[12]. G. U. Selin and I. S. I. Abba, “A Novel Multi ‑ model 10.1007/s11356-021-12792-2.
Data ‑ Driven Ensemble Technique for the Prediction of [23]. G. Elkiran, V. Nourani, and S. I. Abba, “Multi-step
Retention Factor in HPLC Method Development,” ahead modelling of river water quality parameters using
Chromatographia, no. 0123456789, 2020, doi: ensemble artificial intelligence-based approach,” J.
10.1007/s10337-020-03912-0. Hydrol., vol. 577, no. July, p. 123962, 2019, doi:
[13]. H. Varshney, R. A. Khan, U. Khan, and R. Verma, 10.1016/j.jhydrol.2019.123962.
“Approaches of artificial intelligence and machine [24]. V. Nourani, G. Elkiran, and S. I. Abba, “Wastewater
learning in smart cities: Critical review,” IOP Conf. Ser. treatment plant performance analysis using artificial
Mater. Sci. Eng., vol. 1022, no. 1, 2021, doi: intelligence – an ensemble approach,” Water Sci.
10.1088/1757-899X/1022/1/012019. Technol., 2018, doi: 10.2166/wst.2018.477.
[14]. T. Alshammari, N. Alshammari, M. Sedky, and C. [25]. M. S. Gaya, S. I. Abba, A. M. Abdu, and A. I. Tukur,
Howard, “Evaluating Machine Learning Techniques for “Estimation of water quality index using artificial
Activity Classification in Smart Home Environments,” intelligence approaches and multi-linear regression,”
Int. J. Inf. Commun. Eng., vol. 12, no. February, pp. 72– vol. 9, no. 1, pp. 126–134, 2020, doi:
78, 2018. 10.11591/ijai.v9.i1.pp126-134.
[15]. G. Manea, “Green Cities – Urban Planning Models of [26]. G. Usman, S. Işik, and S. I. Abba, “Hybrid data-
the Future,” Cities Glob. world Turkey a Theor. Empir. intelligence algorithms for the simulation of
Perspect., no. November, pp. 462–479, 2014. thymoquinone in Hybrid data ‑ intelligence algorithms
[16]. M. Adamu, S. I. H. Salim, I. Malami, M. N. I. S. I. Abba, for the simulation of thymoquinone in HPLC method
and Y. E. Ibrahim, “Prediction of compressive strength development,” J. Iran. Chem. Soc., no. January, 2021,
of concrete incorporated with jujube seed as partial doi: 10.1007/s13738-020-02124-5.
replacement of coarse aggregate : a feasibility of [27]. R. A. Abdulkadir, S. I. A. Ali, S. I. Abba, and P. Esmaili,
Hammerstein – Wiener model versus support vector “Forecasting of daily rainfall at Ercan Airport Northern
machine,” Model. Earth Syst. Environ., no. Cyprus: a comparison of linear and non-linear models,”
0123456789, 2021, doi: 10.1007/s40808-021-01301-6. Desalin. Water Treat., vol. 177, no. May 2019, pp. 297–
[17]. G. Elkiran, V. Nourani, and S. I. Abba, “Multi-step 305, 2020, doi: 10.5004/dwt.2020.25321.
ahead modelling of river water quality parameters using [28]. M. S. Pankaja, P. D. Student, and H. N. Nagendra,
ensemble artificial intelligence-based approach,” J. “Green City Concept – As New Paradigm in Urban
Hydrol., vol. 577, p. 123962, Oct. 2019, doi: Planning,” Int. J. Eng. Sci., pp. 55–60, 2015.
10.1016/j.jhydrol.2019.123962. [29]. G. Usman, S. Isik, S. I. Abba, and F. Mericli, “Artificial
[18]. M. S. Gaya et al., “Estimation of water quality index intelligence based models for the qualitative and
using artificial intelligence approaches and multi-linear quantitative prediction of a phytochemical compound
regression,” IAES Int. J. Artif. Intell., vol. 9, no. 1, pp. using HPLC method,” Turkish J. Chem., vol. 44, no. 5,
126–134, 2020, doi: 10.11591/ijai.v9.i1.pp126-134. 2021, doi: 10.3906/kim-2003-6.
[19]. M. H. Ahmad, A. G. Usman, and S. I. Abba, [30]. G. Usman, M. H. Ahmad, N. Danraka, and S. I. Abba,
“Comparative performance of extreme learning “The effect of ethanolic leaves extract of
machine and Hammerstein–Weiner models for Hymenodictyon floribundun on inflammatory
modelling the intestinal hyper-motility and secretory biomarkers : a data ‑ driven approach,” Bull. Natl. Res.
inhibitory effects of methanolic leaf extract of Cent., 2021, doi: 10.1186/s42269-021-00586-y.
Combretumhypopilinum Diels (Combretaceae),” Silico [31]. Z. S. A. A.S. Mubarak, Parvaneh Esmaili, M. S. G. ,
Pharmacol., vol. 9, no. 1, 2021, doi: 10.1007/s40203- R.A. Abdulkadir, M. Ozsoz, and S. I. A. , Gaurav Saini,
021-00090-1. “Metro-environmental data approach for the prediction
of chemical oxygen demand in new Nicosia wastewater
treatment plant,” vol. 27049, pp. 1–10, 2021, doi:
10.5004/dwt.2021.27049.

IJISRT21OCT642 www.ijisrt.com 1264

Volume 6, Issue 10, October – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[32]. Musa, N. Yimen, S. I. Abba, H. H. Adun, and M.
Dagbasi, “Multi-state load demand forecasting using
hybridized support vector regression integrated with
optimal design of off-grid energy Systems—a
metaheuristic approach,” Processes, vol. 9, no. 7, 2021,
doi: 10.3390/pr9071166.
[33]. R. Costache et al., “Flash-flood susceptibility
assessment using multi-criteria decision making and
machine learning supported by remote sensing and GIS
techniques,” Remote Sens., vol. 12, no. 1, 2020, doi:
10.3390/RS12010106.
[34]. G. Perboli, A. De Marco, F. Perfetti, and M. Marone, “A
new taxonomy of smart city projects,” Transp. Res.
Procedia, vol. 3, no. July, pp. 470–478, 2014, doi:
10.1016/j.trpro.2014.10.028.
[35]. J. Shuja et al., “Greening emerging IT technologies:
techniques and practices,” J. Internet Serv. Appl., vol. 8,
no. 1, 2017, doi: 10.1186/s13174-017-0060-5.

IJISRT21OCT642 www.ijisrt.com 1265