Future Generation Computer Systems 88 (2018) 696–698

Contents lists available at ScienceDirect

Future Generation Computer Systems

journal homepage: www.elsevier.com/locate/fgcs

Editorial

Recent advances in Big Data Analytics, Internet of Things and Machine

Learning
Roshan Joy Martis a , Varadraj Prabhu Gurupur b , Hong Lin c , Aminul Islam d ,
Steven Lawrence Fernandes e
a
Department of Electronics and Communication Engineering, Vivekananda College of Engineering and Technology, India
b
Department of Health Management and Informatics, University of Central Florida, USA
c
Department of Computer Science & Engineering Technology, University of Houston Downtown, USA
d
School of Computing and Informatics, University of Louisiana at Lafayette, USA
e
Department of Electronics and Communication Engineering, Sahyadri College of Engineering & Management, India

a b s t r a c t

Big data analytics, Internet of Things, and machine learning are some of the rising areas of science and technology forming the next generation of artificial
intelligence-based computing systems. It is also important to note that this aforementioned emerging field is diverse and in some strange ways both
transformative and transdisciplinary in nature. This transformative and transdisciplinary nature of this field enables it to grow both in terms of its
theoretical foundations and applications.
In this special issue we focus on some advanced research projects in these areas that are transformative and transdisciplinary in nature. The projects
and experiments discussed in this special issue constitute the advancement in synthesis of decision support systems that aid further advancement of
healthcare delivery, diagnosing diseases, and analysis of behavioral science.
© 2018 Published by Elsevier B.V.

1. Introduction Knowledge networks and cloud computing have aided the devel-
opment of diagnostic systems, tele-oncology, enabling artificial
Big Data Analytics, Internet of Things, and Machine Learning intelligence in Electronic Health Record Systems, and many other
have a wide range of applications and combined with rich the- such technologies.
oretical foundations. It is worthwhile noting that the theoretical One of our motivations to come up with this special issue was
foundations are transdisciplinary in nature while its applications to bring about innovations that can prove helpful in achieving
are transformative. This transformative and transdisciplinary na- the aforementioned transformative and transdisciplinary objec-
ture of this field [1–5] has made it one of the most coveted for tive. One such example comes from an article published by Xie
scientists and innovators to explore. This exploration ranges from et al., [6]. Here the investigators have successfully applied ad-
medical informatics to prediction of trends in the stock market. The vances in machine learning and Internet of Things to facilitate a
purpose of this special issue is to navigate some of these explo- transformation in healthcare delivery. Another example would be
rations and present it to the readers. It is to be noted that while implementing smart urban planning using Big Data Analytics as
this special issue navigates these aforementioned topics it also described by Babar and Arif [7]. The investigators describing this
points out important information on advancement of theoretical article have used advances in data acquisition, data computation,
foundations in this field; thereby, reflecting the advancement of and decision making formulations in developing their system. It is
science. worthwhile noting that although the application domains for these
This advancement of science is especially true when it comes to two experiments as disparate it is to be noted that they are both
innovative healthcare systems [6]. These innovations have helped transformative in nature.
the scientific community reduce the cost of healthcare while also
Another important contribution worthwhile mentioning will
improving its quality. This transformation in healthcare delivery
be a contribution from Sarwar et al. [8] towards synthesizing a
would not have been possible without the advances in theoret-
research environment for language and literature researchers. In
ical science in the areas of machine learning, data mining, and
this delineated research the investigators have introduced the
knowledge networks. While machine learning and data mining
techniques to manage a variety of data sources for the purpose of
have been around for a few decades the concept of knowledge
eliminating some of the issues associated with data access. Here
networks is slowly but surely emerging as a technology disrup-
we are covering an application domain that has not been covered
tor especially because of its association with cloud computing.
in this section. In any case, the editors would like to mention the
readers of this special issue that there is always room to transform

https://doi.org/10.1016/j.future.2018.07.057
0167-739X/© 2018 Published by Elsevier B.V.
 Editorial / Future Generation Computer Systems 88 (2018) 696–698 697

application domains using the technologies and case studies con- Goyal et al., [17] have described a denoising method from
sidered. morphological filtering in non-sub sampled shearlet transform
domain. The noisy components are processed by morphological
2. Contents of this special issue circular disc operators i.e. Top Hat/ Bottom Hat filtering. The re-
sultant image is decomposed into 8 bit planes and the denoised
In this special issue Bhushan et al., [9] have described the use of image is obtained by combining the bit planes processed with
similarity measure for computing the closeness of two sets of text bitonic filter. Authors claim that this preserves the fine structure
documents. The authors have proposed a compression modeling information when compared to other existing algorithms even at
based on similarity measure between text documents. Two differ- high noise levels.
ent sets of experiments were conducted to validate the efficacy The article titled, ‘‘Socio Evolution & Learning Optimization
of the proposed similarity measures. Experimental results demon- Algorithm: A Socio-inspired Optimization Methodology’’ [18] il-
strate that the f-measure score obtained from similarity metric is lustrate the use of socio-inspired algorithms in identifying man-
better when compared to other text classification techniques. nerisms and behaviors of individuals. The authors claim that they
Magnetic Resonance Angiography is used to detect brain images have used 50 well-known test problems to test their algorithmic
for the purpose of identifying the presence of vascular disorder. approach. This article is full of well described illustrations of the al-
In the article titled [10], ‘‘An approach to examine magnetic reso- gorithms described using good examples involving good diagrams,
nance angiography based on Tsallis entropy and deformable snake scatter plots, and tables.
model’’ the authors provide an improved approach to computer- Pillai et al., [19] in their article titled ‘‘Local Diagonal Extrema
ization of Magnetic Resonance Angiography using Chaotic Firefly Number Pattern: A new Feature Descriptor for Face Recognition’’
algorithm and Tsallis entropy. present an approach which partitions the face into several regions
With the advent of ‘‘Internet of Things’’ many healthcare appli- to facilitate extraction of features from each region individually.
cations are being modernized to improve the accessibility to pa- The extracted features are concatenated into a single feature vector
tients. The article titled, ‘‘Super-Resolution of Retinal Images using which is used as a face descriptor. Only diagonal neighbours are
Multi-Kernel SVR for IoT Healthcare Applications’’ [11] focuses on considered, hence, the dimension of the feature, and the computa-
the use of smartphone fundoscopy for the purpose diagnosing a
tion time to recognize the face are reduced.
patient remotely. This also, ties into the concept of telemedicine
Lin [20] described the use of Autoregressive Conditional Het-
and telehealth.
eroskedastic (ARCH) and Generalized Autoregressive Conditional
The article titled, ‘‘Automated system for the detection of tho-
Heteroskedastic (GARCH) in predicting the volatility of the stock
racolumbar fracture using a CNN architecture’’ [12] illustrate an
markets in China. For this the author used data pertaining to time
automated computer aided diagnosis (CAD) system to accurately
series with associated mathematical formulation to illustrate its
identify the fracture and prescribe correct treatment at the ear-
effective use in implementing time-varying clustering. Addition-
liest. An automated thoracolumbar fracture detection technique
ally, the author also presents variations of these models and its
is proposed using convolutional neural networks (CNNs) without
usage with the purpose of advancing the computational techniques
segmenting the vertebra.
involved.
Bhateja, et al., [13] discuss an improvement in Computer Aided
Khanduzia et al., [21] in their article titled ‘‘Data envelopment
Diagnostic approach for evaluation of cancer by analyzing the
analysis and interdiction median problem with fortification for
mammographic masses. In their article titled, ‘‘Unsharp mask-
enabling IoT technologies to relieve potential attacks’’ present a
ing approaches for HVS based enhancement of mammographic
bi-objective optimization model that optimally serves customers’
masses: a comparative evaluation’’ the authors attempt to describe
a method to automate the identification of the mammographic demands in a remarkably efficient manner and allocates fortifica-
mass using image enhancements. For the purpose of experimen- tion resources to relieve potential disaster, including interdiction
tation, the investigators have analyzed 8 cases to validate their operations. Authors have proposed parking management system
method. The observed qualitative measure indicates a higher per- based on IoT to assist parking drivers in managing and handling
formance compared to other related works. parking in the presence of protection and interdiction operations.
Feature selection is an important upcoming area of research Acharya et al., [22] in their article titled ‘‘Automated identifica-
machine learning, text analysis, and image analysis. Kushwaha, tion of shockable and non-shockable life-threatening ventricular
and Pant [14] in their article titled, ‘‘Link-based BPSO for feature arrhythmias using convolutional neural network’’ present a tool
selection in big data text clustering’’ explain the advancement in that is used to identify shockable and non-shockable ventricular
data mining techniques using their approach. The core objective arrhythmias using neural networks. The tool described in the arti-
investigated is to find a new neighbor selection strategy for fea- cle uses an eleven layer convolutional neural network model. Here
ture extraction. This article uses vector analysis for meeting the we perceive improvement in computational techniques to improve
research objectives and provides explanation using well written healthcare delivery. The editors believe that such contribution
algorithms. in alleviating the present techniques in computational science is
Dogra et al., [15] have designed an image fusion technique to much required.
mitigate the artifact issues in bone and vessel. The source im- The article titled, ‘‘Dynamic Metric Embedding Model for Point-
ages are transformed using KL transformations and Ripplet trans- of-Interest Prediction’’ [23] illustrate a unified flexible model
form. The artifacts are controlled via anisotropic diffusion filtering. which can map temporal (T) and spatial (S) factors into a unified
Laplacian pyramidal based fusion is employed to fuse the mask and Euclidean space and learn them collectively to resolve data spar-
digital subtraction angiography images. sity. Long Short Term Memory is brought into metric embedding
Tan et al., [16] describe the use of convolutional neural network to depict user’s dynamic interest. Furthermore, this is applied to
in their article titled, ‘‘Age-related macular degeneration detection character temporal non-uniformness. Additionally a self-attention
using deep convolutional neural network.’’ This is another article method is developed with the ability to estimate the temporal
that illustrates the use of convolutional neural networks; however, relationship between user’s behaviors.
the case study this time is for problems related to ophthalmology. Overall we believe that this special issue will help many future
The core objective here is to achieve an early diagnosis of Age- projects in the aforementioned areas of research, bolster inno-
related Macular Degeneration. The authors claim that their method vative next generation computing systems, and aid creativity in
has a higher level of accuracy compared to other available methods. the areas of big data analytics, Internet of Things, and machine
698 Editorial / Future Generation Computer Systems 88 (2018) 696–698

learning. It is our perspective that many of these experiments and [11] J. Jebadurai, J.D. Peter, Super-resolution of retinal images using multi-kernel
topics described in this special issue are novel and will assist in SVR for IoT healthcare applications, Future Gener. Comput. Syst. 85 (2018)
338–346.
synthesis and development of future generations of computing
[12] U. Raghavendra, N.S. Bhat, A. Gudigar, U.R. Acharya, Automated system for the
systems. detection of thoracolumbar fractures using a CNN architecture, Future Gener.
Comput. Syst. 85 (2018) 184–189.
References [13] V. Bhateja, M. Misra, S. Urooj, Unsharp masking approaches for HVS based
enhancement of mammographic masses: A comparative evaluation, Future
[1] V. Gurupur, R. Gutierrez, Designing the right framework for healthcare deci- Gener. Comput. Syst. 82 (2018) 176–189.
sion support, J. Integr. Des. Process Sci. (2016). http://dx.doi.org/10.3233/jid- [14] N. Kushwaha, M. Pant, Link based BPSO for feature selection in big data text
2016-0001. clustering, Future Gener. Comput. Syst. 82 (2018) 190–199.
[2] V. Gurupur, M. Tanik, A system for building clinical research applications using [15] A. Dogra, B. Goyal, S. Agrawal, Osseous and digital subtraction angiography
semantic web-based approach, J. Med. Syst. 36 (1) (2012) 53–59. image fusion via various enhancement schemes and Laplacian pyramid trans-
[3] R. Martis, H. Lin, V. Gurupur, S. Fernandes, Editorial: Frontiers in Development formations, Future Gener. Comput. Syst. 82 (2018) 149–157.
of Intelligent Applications for Medical Imaging Processing and Computer Vi- [16] J.H. Tan, S.V. Bhandary, S. Sivaprasad, Y. Hagiwara, A. Bagchi, U. Raghavendra,
sion, Computers in Biology and Medicine. A.K. Rao, B. Raju, N.S. Shetty, A. Gertych, Age-related macular degeneration
[4] S. Fernandes, V. Gurupur, R. Nayak, N. Arunkumar, S. Kadry, A novel nonintru- detection using deep convolutional neural network, Future Gener. Comput.
sive decision support approach for heart rate, Pattern Recognit. Lett. (2017). Syst. 87 (2018) 127–135.
http://dx.doi.org/10.1016/j.patrec.2017.07.002. [17] B. Goyal, A. Dogra, S. Agrawal, B.S. Sohi, Two-dimensional gray scale image
[5] S. Fernandes, B. Chakraborty, V. Gurupur, A. Prabhu, Early skin cancer detection denoising via morphological operations in NSST domain & bitonic filtering,
using computer aided diagnosis techniques, J. Integr. Des. Process Sci. (2016). Future Gener. Comput. Syst. 82 (2018) 158–175.
http://dx.doi.org/10.3233/jid-2016-0002. [18] M. Kumar, A.J. Kulkarni, S.C. Satapathy, Socio evolution & learning optimization
[6] R. Xie, I. Khalil, S. Badsha, M. Atiquzzaman, Fast and peer-to-peer vital signal algorithm: A socio-inspired optimization methodology, Future Gener. Com-
learning system for cloud-based healthcare, Future Gener. Comput. Syst. 88 put. Syst. 81 (2018) 252–272.
(2018) 220–233. [19] A. Pillai, R. Soundrapandiyan, S. Satapathy, S.C. Satapathy, K.-H. Jung, R.
[7] M. Babar, F. Arif, Smart urban planning using big data analytics to contend Krishnan, Local diagonal extrema number pattern: A new feature descriptor
with theInteroperability in Internet of Things, Future Gener. Comput. Syst. 77 for face recognition, Future Gener. Comput. Syst. 81 (2018) 297–306.
(2017) 65–76. [20] Z. Lin, Modelling and forecasting the stock market volatility of SSE Composite
[8] M.S. Sarwar, T. Doherty, J. Watt, R.O. Sinnott, Towards a virtual research Index using GARCH models, Future Gener. Comput. Syst. 79 (3) (2018) 960–
environment for language and literature researchers, Future Gener. Comput. 972.
Syst. 29 (2013) 549–559. [21] R. Khanduzi, M.R. Peyghami, A.K. Sangaiah, Data envelopment analysis and
[9] S.N.B. Bhushan, A. Danti, Classification of compressed and uncompressed text interdiction median problem with fortification for enabling IoT technologies to
documents, Future Gener. Comput. Syst. 88 (2018) 614–623. relieve potential attacks, Future Gener. Comput. Syst. 79 (3) (2018) 928–940.
[10] V. Rajnikanth, N. Dey, S.B. Satapathy, A.S. Ashour, An approach to examine [22] U.R. Acharya, H. Fujita, S.L. Oh, U. Raghavendra, J.H. Tan, M. Adam, A. Gertych,
Magnetic Resonance Angiography based on Tsallis entropy and deformable Y. Hagiwara, Automated identification of shockable and non-shockable life-
snake model, Future Gener. Comput. Syst. 85 (2018) 160–172. threatening ventricular arrhythmias using convolutional neural network,
Future Gener. Comput. Syst. 79 (3) (2018) 952–959.
[23] W. Liu, J. Wang, A.K. Sangaiah, J. Yin, Dynamic metric embedding model for
point-of-interest prediction, Future Gener. Comput. Syst. 83 (2018) 183–192.