International Journal of Computer and Information Technology (ISSN: 2279 – 0764)

Volume 10 – Issue 2, March 2021

Research Methods in Machine Learning: A Content

Analysis
Jackson Kamiri Geoffrey Mariga
School of Computing and Information Technology School of Computing and Information Technology
Murang’a University of Technology Murang’a University of Technology
Murang’a, Kenya Murang’a, Kenya
Email: jkamiri [AT] mut.ac.ke Email: gmariga [AT] mut.ac.ke

Abstract—Research methods in machine learning play a pivotal as scientific research, therefore, it’s research methodology
role since the accuracy and reliability of the results are according to many scholars incorporates several factors
influenced by the research methods used. The main aims of this which include; choice of training data to use, choice of data
paper were to explore current research methods in machine attributes to use, choice of algorithm(s) to use in the model,
learning, emerging themes, and the implications of those themes the ratio between training data and testing data, the
in machine learning research. To achieve this the researchers performance measures of the algorithms, among other
analyzed a total of 100 articles published since 2019 in IEEE factors. Thus, scholars have demonstrated that the choice of
journals. This study revealed that Machine learning uses
quantitative research methods with experimental research
methodology to use should span across the above factors. The
design being the de facto research approach. The study also objectives of the research also influence the methodology to
revealed that researchers nowadays use more than one be used [3]
algorithm to address a problem. Optimal feature selection has
also emerged to be a key thing that researchers are using to Scholars in this field of machine learning have demonstrated
optimize the performance of Machine learning algorithms. that a good research methodology should be capable of
Confusion matrix and its derivatives are still the main ways used clearly explaining how the researcher ended-up with the
to evaluate the performance of algorithms, although researchers results [4]. Therefore, the methodology should be put in such
are now also considering the processing time taken by an a way that another researcher can easily follow and end up
algorithm to execute. Python programming languages together with the same results. Some researchers have opted to use a
with its libraries are the most used tools in creating, training, single approach in terms of choice of the algorithm while
and testing models. The most used algorithms in addressing others use more than one algorithm to model and then they
both classification and prediction problems are; Naïve Bayes,
Support Vector Machine, Random Forest, Artificial Neural
compare the results of the various algorithms [4].
Networks, and Decision Tree. The recurring themes identified Research in the field of machine learning is mainly
in this study are likely to open new frontiers in Machine learning
research.
quantitative since it involves the modeling of data and also
using statistical approaches and formulations to make sense
Keywords: Research methods in machine learning, machine of the data [1], [2], [3], [4]. Therefore, the methods used in
learning algorithms, machine learning techniques. machine learning research in most cases must be consistent
with quantitative research. However, this is not to mean that
I. INTRODUCTION all fields of artificial intelligence or domains affiliated to
Machine learning can be defined as a field of artificial machine learning are confiscated to quantitative research.
intelligence that is concerned with the development of Some areas such as Natural Language Processing may
algorithms and techniques that allow a computer to learn and require a qualitative research approach [5].
gain intelligence from experience [1]. Research methods in
machine learning research play a pivotal role since the When scholars are researching on machine learning they
accuracy and reliability of the results are influenced by the must first scrutinize the problem that they intend to solve and
research approach/ method used. In machine learning, then decide if it requires a quantitative approach, qualitative
models learn from historical data which can either be primary approach, or a hybrid of the two approaches. This is so
data or secondary data. This creates a wide pool of knowledge because machine learning does not exist in a vacuum, it is
from which machines can learn and make decisions based on used to solve problems in society. Thus, the nature of the
what they learn [2]. problem influences the methodology to be used in research
[6]. Also, machine learning is a very active area of research,
The research articles analyzed emphasized the need to choose therefore, scholars are continuously trying to solve a majority
an appropriate methodology. Machine learning is classified

www.ijcit.com 78
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

of the problems in the various domains using machine The apps were assessed against five functions that are
learning techniques. performed by successful caregivers to determine the level to
which the apps relieve the burden of caregivers. The five key
The articles assessed in this paper have shown that research functions were; (i) Information and Resources, (ii) practical
methodology in machine learning can be the key difference problem solving involving behavioral solutions, medication
between various study which are basically solving a similar management, safety, and personal health records tracking. (ii)
problem in the same domain [7,8]. Therefore, the Memory aids, (iv) Family communication which includes
methodology can form the research gap that a researcher notification for checkups, emergency contact list, sharing of
wants to address. The researcher will then use a different critical information. (v) caregiver support that is caring for
approach which s/he thinks is better than the approach(s) used the caregiver. The general assessment of the apps revealed
by other researchers previously to solve the same problem. that 50% (22 apps) were designed for caregivers taking care
Some researchers have also combined methodologies used by of people with memory loss [9].
other researchers with the sole aim of increasing the accuracy
of the results. The results of the analysis against the five functions were as
follows; first, only 34% (15 apps) satisfied the criteria of
providing caregivers with relevant information and resources
Problem statement such as searchable databases [9]. Second, 21 applications met
the criteria for practical problem-solving. Practical problem-
solving in this case was defined as addressing medication
This study aims at exploring research methods in machine management, safety health record tracking, and behavioral
learning and determining emerging themes in the field of solutions [9]. Third, 15 apps passed the test of facilitating
machine learning research. Specifically, the following three better communication and enhancing coordination among
research questions will be addressed. family members [9]. Fourth, 12 applications met the criteria
1) What research methods are commonly used in the of actin as a memory aid. The apps have tools that enable
field of Machine Learning? people with dementia to enhance their cognitive abilities [9].
2) What other recurring themes are there in Machine Five, a total of 10 apps of the 44 apps contained tools that
Learning field with regard to research methods? care for the caregiver by providing emotional support stress
3) What implications do these recurring themes in management, and social support, among others to the
research methods have for the Machine Learning caregivers [9].
discipline? [10] conducted a content analysis of research methods in
Scholars in the field of Machine learning have contributed Library Information Science (LIS). The researcher analyzed
widely to the evolution of research methods. Therefore, a total of 1162 articles published between 2001 and 2010 in
cross-examination of the status quo in the field of machine three major journals of LIS namely; Journal of
learning will inform scholars about research methods and Documentation (JDoc), Journal of the American Society for
help them make more informed decisions while selecting the Information Science, and Technology (JASIS&T), and
methodology to use in their research. Library Information Science research (LIS). However, the
researcher deferred the analysis of articles in the JASIS&T
journal between 2003-2008 and only analyzed articles
II. LITERATURE REVIEW
between 2001-2002 and 2009-2010 [10]. The articles were
A survey of Content Analysis analyzed quantitively and qualitatively to address some
recurring themes about research method selection and
Grossman, Zak, and Zelinsk conducted a quantitative content application in the scholarly domain. Non-research articles in
analysis on mobile applications used by caregivers. In the this field were excluded from the analysis.
analysis, 44 applications that self-identified themselves or
were advertised as caregivers aiding tool were assessed. The The findings of the research were as follows: the three
apps were identified in two major application stores namely journals shared 4 of the top five research methods identified
iTunes App Store and Google play store. To get the apps, the in this study. The theoretical approach leads with a
researchers searched the keywords “caregivers”, cumulative percentage of 65%, the content analysis took the
“caregiving” and “elderly care” [9]. The researchers only second position with a percentage of 57%, Questionnaire
include applications that exclusively dealt with informal or becomes position three with a cumulative percentage of
family caregivers for elderly people. Therefore, apps 55.8%, and experiment occupied position four with a 53.4%
developed for professional caregivers, and those developed cumulative percentage score [10]. The researcher discovered
for people with special chronic conditions such as diabetes that unlike before, questionnaires and surveys no longer
were excluded [9]. dominate as the leading research methods in Library
Information Science [10]. Despite Bibliometrics leading in
JASIS&T, it did not feature well in the other journals,

www.ijcit.com 79
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

therefore, it did not qualify to be among the top 4 techniques. III. METHOD
The researcher further reports that the use of multiple One hundred research articles published since 2019 were
methods has gained a lot of prevalence. obtained from IEEE journals. All the research papers used
were in the field of machine learning. Journal papers that
Azeez and Van der Vyver conducted a comprehensive were not written in English or at least translated to English
content analysis on privacy and security issues in e-health were excluded. Access to these journals was provided by
cloud-based model. The scholar reviewed 110 articles in this Murang’a University of Technology.
area and discovered several models discovered in their The data collection yielded 100 research articles in the field
solutions. The articles were sourced from the following of machine learning as summarized on table1. These articles
journals; ACM digital library, IEEE digital library, IEE were considered to be very relevant since they were not more
explore digital library, Springer, Elsevier, and Science direct than 2 years old, thus they can inform current research in the
[11]. The 110 reviewed articles were arrived at after field of machine learning which is a very active field. The
comparing the models and approaches used by many research articles chosen were solving different problems
researchers. The researchers examined the strengths and using machine learning approaches. All the 100 articles were
weaknesses of each method adopted in addressing E-health analyzed by the researchers to determine the research
security challenges [11]. methods used in machine learning.
The scholars discovered that for any E-health system to be In this study, the researchers analyzed each of the sampled
reliable it must have strong mechanisms to counter various papers along several dimensions to describe machine
security threats such as; tampering, masquerading, denial of learning research methods. These dimensions are: General
service attack, and privacy [11]. Public Key Encryption was research design, sources of data, algorithms used, and
identified as the most widely used security mechanism, approaches applied in data pre-processing, model training,
however, the researcher record that it needs to be enhanced. testing, and evaluation. The above issues were used to inform
ABE is another security scheme that is being used. ABE is the findings of this paper. All the 100 research papers used in
efficient in enhancing security and privacy although it is this study had the
expensive considering computational complexity. The high
cost of ABE is mainly because the ciphertext is expensive above key parameters in their research methodology. These
thus decryption is expensive [11]. parameters are just general parameters, therefore, under each
one of them, several sub-parameters were identified. For
Selective encryption was also widely used in the articles example, under the performance metrics, the majority of the
assed. This involves encrypting the parts of the data that are articles recorded the performance of the algorithms used
considered to be more sensitive [11]. This method is under the various performance metrics.
considered to be less costly, therefore, researchers are
considering it over ABE. The researchers concluded that
research in security and privacy issues in E-health is still
wanting since threats are increasing with time.
TABLE 1: RESEARCH ARTICLES ANALYZED

ID Application Area Number of Articles References of Analyzed Articles

1 Machine Learning Algorithm 5 [16], [85], [88], [89], [95]
Enhancement
2 Solutions in Engineering 8 [14], [27], [62], [76], [86], [92], [110], [114]
3 Solutions in computer networks and 23 [17], [40], [49], [50], [51], [63], [64], [65], [68],
security [70], [72], [80], [93], [94], [99],
[101], [103], [104], [105], [106], [113], [120], [122],
[123]
4 Natural Language Processing 8 [28], [20], [44], [47], [74], [96], [98], [112]
5 Image Processing 10 [29], [32], [37], [55], [69], [73], [77], [84], [87],
[119]
6 Solutions Health 24 [1], [30], [31], [19], [34], [23], [43], [46], [48], [53],
[57], [61], [75], [78], [81], [82], [83], [90], [97],
[111], [115], [116], [118], [124]
7 Solutions in Software Engineering 1 [39]
8 Solutions in Internet of Things 5 [41], [54], [91], [100], [121]
9 Solutions in Education 2 [42,45]
10 Solutions in Analytics and Artificial 13 [52], [3], [58], [59], [60], [66], [67], [79], [102],
Intelligence [107], [108], [109], [117]

www.ijcit.com 80
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

Table 1 shows a summary of research articles analyzed in this repository was the UCI ML repository. This repository has
research. The table portrays that machine learning techniques been used by research works in machine learning that are
are being are applied to address problems in diverse fields. addressing diverse problems. This shows that the repository
has a variety of datasets in different areas. Apart from
repositories, some research articles used data obtained from
IV. RESULTS government entities and Non-governmental organizations.

This section presents results of data analysis along the

dimensions indicated in section 1.3.

General Research Design 20%

Findings of this study indicates that research in machine
learning is mainly quantitative. Therefore, it has standardized
approaches and uses statistical models in the various
segments of the research. All articles analyzed in this study
used an experimental research design. This involved the
design of an experiment and conducting the experiment to 80%
obtain results. The general approach used in conducting
experiments in all the analyzed research articles is as follows:
Data collection, Data pre-processing, Model training, model
testing, model evaluation. These steps are summarized in
Figure 1. Secondary data Only Primary + Secondary Data

Figure 2: Type of Data Used

Also, of the researchers who used secondary data, 87% used

a single data set while 13% used more than one data sets for
example [17] has used three datasets as summarized in Figure
3. Among the researchers who used more than one dataset,
the majority sourced some data set from repositories and
some of the data set from either governmental or non-
governmental organizations. Others sourced all their data sets
from repositories. The key reason for using more than one
dataset according to the researchers was to enhance the
validity and reliability of the model [17].

13%

Figure 1: General Research Approach 87%

Single Data set More than One Data set

Data Collection
Figure 3: Secondary data usage analysis
In this stage, two main sources of data were used in the
articles analyzed. The two sources are secondary data and Primary data was mainly sourced from recording data over a
primary data. 80% of the articles used purely secondary data long period. Research that involved primary data took a
sets in the research as shown in Figure 2. The secondary data longer time than those that used secondary data. A
set was obtained from data repositories. The main used data combination of both primary and secondary data was used

www.ijcit.com 81
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

where there is a need to guarantee the validity of the model Data normalization Involved tasks such as formatting the data
or to enhance the model to accommodate real-time data [12]. in a form that is easier for the algorithms to understand during
Also, where secondary data is not available or the available training. It also involved data scaling and ensuring that the
secondary data is not sufficient then the researchers resulted training data is in the same format as the testing data. It was
in using either purely primary data or a combination of both noted that normalized data promotes the efficiency of
primary and secondary data. machine learning models [17].
None of the articles analyzed in this study used more than one Noise reduction involved the elimination of outliers or any
set of primary data. Also, none of the research works used data item that is inconsistent with the other parts of the data.
qualitative data collection methods to collect any form of Optimal feature selection involved the selection of critical
primary data. This, therefore, proves that research in machine features that influence the results also known as the
learning is purely quantitative. dependent variable in a great way. Articles that applied this
technique first modeled the data without extracting optimal
Data Pre-processing features and then in the second phase the optimal features
were selected and the data modeled again.
All the articles analyzed had this stage as part of their
methodology. The majority of the articles noted that this is a In all the articles that used optimal feature extraction, it was
very critical stage in machine learning since this stage can noted that the performance of the models in terms of accuracy
influence the performance of models and consequently and also in the confusion matrix improved after optimal
influence the results of the model. This is so because machine features were selected [13], [14], [15], [16]. This is a new
learning algorithms learn from the data provided to them and development in the field of machine learning that is attracting
use the acquired knowledge to make decisions. a lot of interest among researchers since it improves
performance.
The analyzed articles revealed that despite the algorithms to
be used in the research (supervised, unsupervised, or Principle component analysis PCA has been used as one of
reinforcement learning) or the nature of the research the major techniques of selecting optimal features [13], [14].
(classification or regression) data pre-processing is a It identifies correlation among various attributes of a dataset,
paramount stage that cannot be omitted. Some articles therefore, enabling researchers to determine how each
recorded that much focus in their research work was on data attribute influences the results and the relevance of such
pre-processing. attribute in that particular data set [14].

Data pre-processing involved a series of activities. The Choice of algorithms

activities that were unanimous among all analyzed research
articles were; data cleanup, data normalization, and noise All the articles analyzed used at least one machine-learning
reduction e.g. [17], [18]. 10% of total analyzed articles algorithm to solve the research problem that they were
performed feature extraction/ optimal feature selection in addressing. It was noted that the choice of algorithm depends
data pre-processing as illustrated in Figure 4. Data cleaning mainly on the nature of the problem to be solved. According
involved activities such as filling-in null values or to the analyzed articles, problems can be grouped as either
eliminating null values if they cannot be filled. classification problems or regression/ prediction problems.
Also, problems can be classified as either supervised learning
or unsupervised learning problems.

10% Some Machine learning algorithms can be used to solve

classification problems only while others can be used to solve
regression problems only, while others solve both
classification and regression problems. It was noted that
algorithms that solve both classification and regression
problems have attracted a lot of favor from the researchers as
90% per the articles analyzed.
The main pure classification algorithm used are K-Nearest
Neighbors and Logistic regression. The main purely
Performed feature Selection prediction algorithm used is Linear regression. The most
commonly used algorithms that support both classification
Did not Perform Feature Selection and prediction are; Naïve Bayes, Support Vector Machine,
Random Forest, Artificial Neural Networks, and Decision
Figure 4: Data Pre-processing Tree as illustrated in Figure 5.

www.ijcit.com 82
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

The analyzed articles revealed that some algorithms such as

Neural networks algorithms may take more processing time
when dealing with large sets of data but will generate better
results in terms of accuracy. Others like Decision trees
perform better with small sets of data; however, the decision
tree can be unstable. Support vector machine can handle
complex functions and even scale up with dimensional data,
however, it is not suitable with outliers [21].
The above critical features of algorithms influenced their
performance and thus the nature of the data used determined
which algorithm will register the highest performance. It was
discovered that by using multiple algorithms, researchers can
discover the most suitable algorithm for a given problem.
Model Training and Testing
Figure 5: Main Algorithms used
All the analyzed articles included model training as part of
It was noted that the above algorithms which support both
their methodology. 80% of the articles analyzed in this study
classification and prediction have modules within them that
used secondary data to train the model of these, 83 % used a
perform a specific task. For example, it was identified that
single data set for both training and testing. The data set was
Convolutional Neural Networks is part of Artificial Neural
broken down into a training set and a testing set. The majority
Networks that deals with image classification [19]. This
used 70% of the data for training while the remaining 30%
enables the algorithms to perform numerous tasks in both
was used for testing. Others used 75% of the data for training
classification and regression. Among the analyzed articles,
and 25% of the data was reserved for testing.
53% have specified the specific element of such algorithms
that they used in their study. This is a new paradigm shift 20% of the total articles analyzed used 10-fold cross-
where researchers are now not just concerned about the validation to train and test the model. In this approach, the
algorithm to use in solving a problem but also the specific data set was divided into 10 equal parts. 9 parts or segments
module of the algorithm that if used will generate optimal were used to train the model and then the remaining 1
results [20]. segment was used to test the model. This was repeated 10
times with the testing segment being alternated among the 10
98% of the analyzed articles used more than one algorithm to
equal parts. Therefore, each part was used for testing once
model as illustrated in Figure 6. Some used as many as five
and for training 9 times.
algorithms to model the data and then compared the results
of the various algorithms. It was noted that this was aimed at 13% of the researchers who used secondary data used
increasing the reliability of results obtained since in using different data sets for training and testing. Some used two
more than one algorithm, the results become immune to the data sets in their research. One set was used for training while
limitations of any particular algorithms. the other set was used for testing the model. Some researchers
though very few used secondary data to train the model and
2% primary data to test the model.
It was discovered that researchers who used different sets to
train and test their models had to do a lot of data pre-
processing to ensure that optimal features in both training and
testing data sets are well identified and also the attributes are
properly labeled so that the model does not get confused.

98% Model Evaluation

The evaluation matrix that was used by all the researchers in

the articles analyzed is the confusion matrix. The researchers
One More than One who used more than one algorithm evaluated the performance
of each algorithm in the confusion matrix and its derivatives.
Figure 6: Number of algorithms used The other measure that 40% of the researchers used on top of
the confusion matrix is the processing time taken by each
algorithm as shown in Figure 7.

www.ijcit.com 83
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

and test the model. 1% used a combination of machine

learning approaches implemented using python and Fuzzy
logic implemented in MATLAB.
40%
10% 1%
60%

Confusion Matrix only

Confusion Matrix+ Algorithm processing time
89%
Figure 7: Model Evaluation
Python R Python+Fuzzy logic
The confusion matrix has the following parameters which
were used by the researchers to come up with the derivatives Figure 8:Programming Techniques Used
of the confusion matrix [22].
This shows that Python programming language and its
a) True Positive (TP) libraries are gaining a lot of preference among researchers in
b) True Negative (TN) the field of machine learning as opposed to R. Python is
c) False Positive (FP) delivering an easier and more efficient and effective way of
d) False Negative (FN) doing machine learning research. Also, the vast libraries in
The derivatives of these matrices are [22,23]; python help researchers to perform numerous activities on the
data or models with a lot of conveniences.
a) Accuracy- (TP+TN) / (TP+TN+FP+FN)
b) Precision= (TP)/ (TP+FP) V. DISCUSSION
c) Recall= (TP)/ (TP+FN) This study has discovered that research in Machine learning
d) F-Score= 2* (precision *Recall)/ (Recall+ is attracting a lot of interest among researchers which is a
Precision) clear demonstration that Machine Learning is a very active
area of research.
Evaluation of algorithms against the above parameters
generated a true picture of how the algorithm has performed. The most recurring themes in machine learning research
Some researchers were interested in only the accuracy of the include;
algorithm while others were concerned about the
performance of algorithms against all the parameters of the i. The use of algorithms that can perform both
matrix. classification and prediction
ii. The use of more than one algorithm to address a
It was discovered that some researchers considered running problem to enhance reliability
the models more than once and recorded the performance of iii. Optimal feature selection in data pre-processing as
each of the test runs, then computed the average performance. a way of improving performance
It was also discovered that the researchers who considered iv. The use of specific modules of an algorithm to
processing time as a measure of performance, went a step address a problem in more effective and efficient
further and evaluated processing time against the accuracy of ways.
each algorithm in their research to determine the most v. The use of both processing time and confusion
optimal algorithms. matrix derivatives to evaluate machine learning
models.
One new approach that was used to evaluate the processing
time against the accuracy of each algorithm was the use of These recurring themes are likely to open new frontiers in
regression equations X and Y [24]. machine learning research. Also, due to the active nature of
this field of research, more techniques are likely to pop-up
Tools and Techniques used with time.
89% of the articles analyzed used python programming
language to develop the model, train the model, and test the
model. The most utilized python library is the Scikit-Learn
library. 10% of the analyzed articles used R to develop, train,

www.ijcit.com 84
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

VI. CONCLUSION AND FUTURE WORK Hefei, China, 2020, pp. 1446-1450, doi:
10.1109/CCDC49329.2020.9164457.
The key contribution of this study is that it not only provides [8] M. R. Ullah, N. A. Dola, A. Sattar and A. Hasnat,
an updated view of research methods in the field of Machine "Plant Diseases Recognition Using Machine Learning," 2019
Learning but also discusses emerging issues that are worth 8th International Conference System Modeling and
the attention of researchers in the field of machine learning. Advancement in Research Trends (SMART), Moradabad,
The findings of the study should help Machine Learning India, 2019, pp. 67-73, doi:
researchers to make informed decisions while pursing 10.1109/SMART46866.2019.9117284.
research in machine learning. [9] M. R. Grossman, D. K. Zak, and E. M. Zelinski,
This study has focused more on general research methods in “Mobile apps for caregivers of older adults: Quantitative
machine learning. Future researchers can perform a content content analysis.” 2018 JMIR mHealth and uHealth, 6(7),
analysis of specific areas of machine learning such as e162.
supervised learning, unsupervised learning, video analytics, [10] H. Ch. “Research methods in library and
text analytics, classification, and prediction. information science: A content analysis.” 2015 Library &
information science research, 37(1), 36-41.
REFERENCES [11] N. A. Azeez, & C. Van der Vyver. “Security and
[1] H. Kaur and V. Kumari, “Predictive modelling and privacy issues in e-health cloud-based system: A
analytics for diabetes using a machine learning approach”, comprehensive content analysis.” 2019 Egyptian Informatics
2020 Applied computing and informatics. Journal, 20(2), 97-108.
[2] X. Li, K. Li, D. Qiao, Y. Ding and D. Wei, [12] O. Amale and R. Patil, "“IOT Based Rainfall
"Application Research of Machine Learning Method Based Monitoring System Using WSN Enabled Architecture”,"
on Distributed Cluster in Information Retrieval," 2019 2019 3rd International Conference on Computing
International Conference on Communications, Information Methodologies and Communication (ICCMC), Erode, India,
System and Computer Engineering (CISCE), Haikou, China, 2019, pp. 789-791, doi: 10.1109/ICCMC.2019.8819721.
2019, pp. 411-414, doi: 10.1109/CISCE.2019.00097. [13] A. Rehman, A. Khan, M. A. Ali, M. U. Khan, S. U.
[3] K. Pahwa and N. Agarwal, "Stock Market Analysis Khan and L. Ali, "Performance Analysis of PCA, Sparse
using Supervised Machine Learning," 2019 International PCA, Kernel PCA and Incremental PCA Algorithms for
Conference on Machine Learning, Big Data, Cloud and Heart Failure Prediction," 2020 International Conference on
Parallel Computing (COMITCon), Faridabad, India, 2019, Electrical, Communication, and Computer Engineering
pp. 197-200, doi: 10.1109/COMITCon.2019.8862225. (ICECCE), Istanbul, Turkey, 2020, pp. 1-5, doi:
[4] K. A. Taher, B. Mohammed Yasin Jisan and M. M. 10.1109/ICECCE49384.2020.9179199.
Rahman, "Network Intrusion Detection using Supervised [14] K. Chen, "Indirect PCA Dimensionality Reduction
Machine Learning Technique with Feature Selection," 2019 Based Machine Learning Algorithms for Power System
International Conference on Robotics,Electrical and Signal Transient Stability Assessment," 2019 IEEE Innovative
Processing Techniques (ICREST), Dhaka, Bangladesh, 2019, Smart Grid Technologies - Asia (ISGT Asia), Chengdu,
pp. 643-646, doi: 10.1109/ICREST.2019.8644161. China, 2019, pp. 4175-4179, doi: 10.1109/ISGT-
[5] Rahul, S. Adhikari and Monika, "NLP based Asia.2019.8881370.
Machine Learning Approaches for Text Summarization," [15] P. Sun, D. Wang, V. C. Mok and L. Shi,
2020 Fourth International Conference on Computing "Comparison of Feature Selection Methods and Machine
Methodologies and Communication (ICCMC), Erode, India, Learning Classifiers for Radiomics Analysis in Glioma
2020, pp. 535-538, doi: Grading," in IEEE Access, vol. 7, pp. 102010-102020, 2019,
10.1109/ICCMC48092.2020.ICCMC-00099. doi: 10.1109/ACCESS.2019.2928975.
[6] I. C. Freeman, A. J. Haigler, S. E. Schmeelk, L. R. [16] P. Nair and I. Kashyap, "Hybrid Pre-processing
Ellrodt and T. L. Fields, "What are they Researching? Technique for Handling Imbalanced Data and Detecting
Examining Industry-Based Doctoral Dissertation Research Outliers for KNN Classifier," 2019 International Conference
through the Lens of Machine Learning," 2018 17th IEEE on Machine Learning, Big Data, Cloud and Parallel
International Conference on Machine Learning and Computing (COMITCon), Faridabad, India, 2019, pp. 460-
Applications (ICMLA), Orlando, FL, 2018, pp. 1338-1340, 464, doi: 10.1109/COMITCon.2019.8862250.
doi: 10.1109/ICMLA.2018.00217. [17] P. Raghavan and N. E. Gayar, "Fraud Detection
[7] D. Jiang, F. Li, Y. Yang and S. Yu, "A Tomato Leaf using Machine Learning and Deep Learning," 2019
Diseases Classification Method Based on Deep Learning," International Conference on Computational Intelligence and
2020 Chinese Control And Decision Conference (CCDC), Knowledge Economy (ICCIKE), Dubai, United Arab

www.ijcit.com 85
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

Emirates, 2019, pp. 334-339, doi: Conference on Data Stream Mining & Processing (DSMP),
10.1109/ICCIKE47802.2019.9004231. Lviv, Ukraine, 2020, pp. 116-120, doi:
[18] A. Juneja and N. N. Das, "Big Data Quality 10.1109/DSMP47368.2020.9204036.
Framework: Pre-Processing Data in Weather Monitoring [27] N. A. J. Sufri, N. A. Rahmad, N. F. Ghazali, N.
Application," 2019 International Conference on Machine Shahar and M. A. As’ari, "Vision Based System for Banknote
Learning, Big Data, Cloud and Parallel Computing Recognition Using Different Machine Learning and Deep
(COMITCon), Faridabad, India, 2019, pp. 559-563, doi: Learning Approach," 2019 IEEE 10th Control and System
10.1109/COMITCon.2019.8862267. Graduate Research Colloquium (ICSGRC), Shah Alam,
[19] S. Hussein, P. Kandel, C. W. Bolan, M. B. Wallace Malaysia, 2019, pp. 5-8, doi:
and U. Bagci, "Lung and Pancreatic Tumor Characterization 10.1109/ICSGRC.2019.8837068.
in the Deep Learning Era: Novel Supervised and [28] R. Katarya and P. Srinivas, "Predicting Heart
Unsupervised Learning Approaches," in IEEE Transactions Disease at Early Stages using Machine Learning: A Survey,"
on Medical Imaging, vol. 38, no. 8, pp. 1777-1787, Aug. 2020 International Conference on Electronics and
2019, doi: 10.1109/TMI.2019.2894349. Sustainable Communication Systems (ICESC), Coimbatore,
[20] S. Lim and H. Han, "Distinguishing Political Parties India, 2020, pp. 302-305, doi:
From Inaugural Speeches Using Machine Learning," 2020 10.1109/ICESC48915.2020.9155586.
IEEE Conference on Multimedia Information Processing and [29] S. Mojrian et al., "Hybrid Machine Learning Model
Retrieval (MIPR), Shenzhen, Guangdong, China, 2020, pp. of Extreme Learning Machine Radial basis function for
97-100, doi: 10.1109/MIPR49039.2020.00026. Breast Cancer Detection and Diagnosis; a Multilayer Fuzzy
[21] S. Ray, "A Quick Review of Machine Learning Expert System," 2020 RIVF International Conference on
Algorithms," 2019 International Conference on Machine Computing and Communication Technologies (RIVF), Ho
Learning, Big Data, Cloud and Parallel Computing Chi Minh City, Vietnam, 2020, pp. 1-7, doi:
(COMITCon), Faridabad, India, 2019, pp. 35-39, doi: 10.1109/RIVF48685.2020.9140744.
10.1109/COMITCon.2019.8862451. [30] P. K. Fahira, Z. P. Rahmadhani, P. Mursanto, A.
[22] A. Bhatnagar and S. Srivastava, "A Robust Model Wibisono and H. A. Wisesa, "Classical Machine Learning
for Churn Prediction using Supervised Machine Learning," Classification for Javanese Traditional Food Image," 2020
2019 IEEE 9th International Conference on Advanced 4th International Conference on Informatics and
Computing (IACC), Tiruchirappalli, India, 2019, pp. 45-49, Computational Sciences (ICICoS), Semarang, Indonesia,
doi: 10.1109/IACC48062.2019.8971494. 2020, pp. 1-5, doi: 10.1109/ICICoS51170.2020.9299039.
[23] P. Sonar and K. JayaMalini, "Diabetes Prediction [31] N. Reamaroon, M. W. Sjoding, K. Lin, T. J.
Using Different Machine Learning Approaches," 2019 3rd Iwashyna and K. Najarian, "Accounting for Label
International Conference on Computing Methodologies and Uncertainty in Machine Learning for Detection of Acute
Communication (ICCMC), Erode, India, 2019, pp. 367-371, Respiratory Distress Syndrome," in IEEE Journal of
doi: 10.1109/ICCMC.2019.8819841. Biomedical and Health Informatics, vol. 23, no. 1, pp. 407-
[24] L. M. B. Alonzo, F. B. Chioson, H. S. Co, N. T. 415, Jan. 2019, doi: 10.1109/JBHI.2018.2810820.
Bugtai and R. G. Baldovino, "A Machine Learning Approach [32] G. A. Tahir and C. K. Loo, "An Open-Ended
for Coconut Sugar Quality Assessment and Prediction," 2018 Continual Learning for Food Recognition Using Class
IEEE 10th International Conference on Humanoid, Incremental Extreme Learning Machines," in IEEE Access,
Nanotechnology, Information Technology, Communication vol. 8, pp. 82328-82346, 2020, doi:
and Control, Environment and Management (HNICEM), 10.1109/ACCESS.2020.2991810.
Baguio City, Philippines, 2018, pp. 1-4, doi: [33] R. Singh, J. Singh, M. S. Gill, R. Malhotra and
10.1109/HNICEM.2018.8666315. Garima, "Transfer Learning Code Vectorizer based Machine
[25] A. Lasisi, M. O. Sadiq, I. Balogun, A. Tunde-Lawal Learning Models for Software Defect Prediction," 2020
and N. Attoh-Okine, "A Boosted Tree Machine Learning International Conference on Computational Performance
Alternative to Predictive Evaluation of Nondestructive Evaluation (ComPE), Shillong, India, 2020, pp. 497-502, doi:
Concrete Compressive Strength," 2019 18th IEEE 10.1109/ComPE49325.2020.9200076.
International Conference On Machine Learning And [34] K. A. Taher, B. Mohammed Yasin Jisan and M. M.
Applications (ICMLA), Boca Raton, FL, USA, 2019, pp. Rahman, "Network Intrusion Detection using Supervised
321-324, doi: 10.1109/ICMLA.2019.00060. Machine Learning Technique with Feature Selection," 2019
[26] J. Rasheed, H. B. Dogru and A. Jamil, "Turkish Text International Conference on Robotics,Electrical and Signal
Detection System from Videos Using Machine Learning and Processing Techniques (ICREST), Dhaka, Bangladesh, 2019,
Deep Learning Techniques," 2020 IEEE Third International pp. 643-646, doi: 10.1109/ICREST.2019.8644161.

www.ijcit.com 86
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

[35] K. N. Bhanu, H. J. Jasmine and H. S. [44] A. Halimaa A. and K. Sundarakantham, "Machine

Mahadevaswamy, "Machine learning Implementation in IoT Learning Based Intrusion Detection System," 2019 3rd
based Intelligent System for Agriculture," 2020 International International Conference on Trends in Electronics and
Conference for Emerging Technology (INCET), Belgaum, Informatics (ICOEI), Tirunelveli, India, 2019, pp. 916-920,
India, 2020, pp. 1-5, doi: doi: 10.1109/ICOEI.2019.8862784.
10.1109/INCET49848.2020.9153978. [45] H. M. E. Misilmani and T. Naous, "Machine
[36] T. Toivonen, I. Jormanainen, J. Kahila, M. Tedre, T. Learning in Antenna Design: An Overview on Machine
Valtonen and H. Vartiainen, "Co-Designing Machine Learning Concept and Algorithms," 2019 International
Learning Apps in K–12 With Primary School Children," Conference on High Performance Computing & Simulation
2020 IEEE 20th International Conference on Advanced (HPCS), Dublin, Ireland, 2019, pp. 600-607, doi:
Learning Technologies (ICALT), Tartu, Estonia, 2020, pp. 10.1109/HPCS48598.2019.9188224.
308-310, doi: 10.1109/ICALT49669.2020.00099. [46] K. Al-Gumaei, A. Müller, J. N. Weskamp, C. S.
[37] T. Mehejabin, F. Rahman, S. Yeasmin and M. Longo, F. Pethig and S. Windmann, "Scalable Analytics
Sarkar, "Identification of Most Relevant Breast Cancer Platform for Machine Learning in Smart Production
miRNA using Machine Learning Algorithms," 2020 11th Systems," 2019 24th IEEE International Conference on
International Conference on Computing, Communication and Emerging Technologies and Factory Automation (ETFA),
Networking Technologies (ICCCNT), Kharagpur, India, Zaragoza, Spain, 2019, pp. 1155-1162, doi:
2020, pp. 1-6, doi: 10.1109/ICCCNT49239.2020.9225624. 10.1109/ETFA.2019.8869075.
[38] K. Indira, E. Brumancia, P. S. Kumar and S. P. T. [47] E. A. Bayrak, P. Kırcı and T. Ensari, "Comparison
Reddy, "Location prediction on Twitter using machine of Machine Learning Methods for Breast Cancer Diagnosis,"
learning Techniques," 2019 3rd International Conference on 2019 Scientific Meeting on Electrical-Electronics &
Trends in Electronics and Informatics (ICOEI), Tirunelveli, Biomedical Engineering and Computer Science (EBBT),
India, 2019, pp. 700-703, doi: Istanbul, Turkey, 2019, pp. 1-3, doi:
10.1109/ICOEI.2019.8862768. 10.1109/EBBT.2019.8741990.
[39] R. Farhat, Y. Mourali, M. Jemni and H. Ezzedine, [48] A. Mujib and T. Djatna, "Ensemble Learning for
"An overview of Machine Learning Technologies and their Predictive Maintenance on Wafer Stick Machine Using IoT
use in E-learning," 2020 International Multi-Conference on: Sensor Data," 2020 International Conference on Computer
“Organization of Knowledge and Advanced Technologies” Science and Its Application in Agriculture (ICOSICA),
(OCTA), Tunis, Tunisia, 2020, pp. 1-4, doi: Bogor, Indonesia, 2020, pp. 1-5, doi:
10.1109/OCTA49274.2020.9151758. 10.1109/ICOSICA49951.2020.9243180.
[40] M. M. Baig et al., "Machine Learning-based Risk of [49] S. Mirzaei, T. Sidi, C. Keasar and S. Crivelli,
Hospital Readmissions: Predicting Acute Readmissions "Purely Structural Protein Scoring Functions Using Support
within 30 Days of Discharge," 2019 41st Annual Vector Machine and Ensemble Learning," in IEEE/ACM
International Conference of the IEEE Engineering in Transactions on Computational Biology and Bioinformatics,
Medicine and Biology Society (EMBC), Berlin, Germany, vol. 16, no. 5, pp. 1515-1523, 1 Sept.-Oct. 2019, doi:
2019, pp. 2178-2181, doi: 10.1109/EMBC.2019.8856646. 10.1109/TCBB.2016.2602269..
[41] S. Satpathy, S. P. Mishra and A. K. Nayak, [50] T. Badriyah, N. Sakinah, I. Syarif and D. R. Syarif,
"Analysis of Learning Approaches for Machine Translation "Machine Learning Algorithm for Stroke Disease
Systems," 2019 International Conference on Applied Classification," 2020 International Conference on Electrical,
Machine Learning (ICAML), Bhubaneswar, India, 2019, pp. Communication, and Computer Engineering (ICECCE),
160-164, doi: 10.1109/ICAML48257.2019.00038. Istanbul, Turkey, 2020, pp. 1-5, doi:
[42] R. Tao et al., "Magnetocardiography-Based 10.1109/ICECCE49384.2020.9179307.
Ischemic Heart Disease Detection and Localization Using [51] P. S. Gouripeddi, R. Gouripeddi and S. P.
Machine Learning Methods," in IEEE Transactions on Gouripeddi, "Toward Machine Learning and Big Data
Biomedical Engineering, vol. 66, no. 6, pp. 1658-1667, June Approaches for Learning Analytics," 2019 IEEE Tenth
2019, doi: 10.1109/TBME.2018.2877649. International Conference on Technology for Education
[43] R. Trinchero and F. G. Canavero, "Machine (T4E), Goa, India, 2019, pp. 256-257, doi:
Learning for the Design of a Distribution Network for High- 10.1109/T4E.2019.000-6.
Speed Signals," 2019 International Conference on [52] G. Meena, D. Sharma and M. Mahrishi, "Traffic
Electromagnetics in Advanced Applications (ICEAA), Prediction for Intelligent Transportation System using
Granada, Spain, 2019, pp. 1038-1041, doi: Machine Learning," 2020 3rd International Conference on
10.1109/ICEAA.2019.8879415. Emerging Technologies in Computer Engineering: Machine

www.ijcit.com 87
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

Learning and Internet of Things (ICETCE), Jaipur, India, Telecommunication Systems (CITS), Beijing, China, 2019,
2020, pp. 145-148, doi: pp. 1-5, doi: 10.1109/CITS.2019.8862098.
10.1109/ICETCE48199.2020.9091758. [62] A. Asokan and J. Anitha, "Machine Learning based
[53] P. Chopade, D. Edwards, S. M. Khan, A. Andrade Image Processing Techniques for Satellite Image Analysis -
and S. Pu, "CPSX: Using AI-Machine Learning for Mapping A Survey," 2019 International Conference on Machine
Human-Human Interaction and Measurement of CPS Learning, Big Data, Cloud and Parallel Computing
Teamwork Skills," 2019 IEEE International Symposium on (COMITCon), Faridabad, India, 2019, pp. 119-124, doi:
Technologies for Homeland Security (HST), Woburn, MA, 10.1109/COMITCon.2019.8862452.
USA, 2019, pp. 1-6, doi: 10.1109/HST47167.2019.9032906. [63] R. Vishwakarma and A. K. Jain, "A Honeypot with
[54] M. R. Ahmed, Y. Zhang, Z. Feng, B. Lo, O. T. Inan Machine Learning based Detection Framework for defending
and H. Liao, "Neuroimaging and Machine Learning for IoT based Botnet DDoS Attacks," 2019 3rd International
Dementia Diagnosis: Recent Advancements and Future Conference on Trends in Electronics and Informatics
Prospects," in IEEE Reviews in Biomedical Engineering, vol. (ICOEI), Tirunelveli, India, 2019, pp. 1019-1024, doi:
12, pp. 19-33, 2019, doi: 10.1109/RBME.2018.2886237. 10.1109/ICOEI.2019.8862720.
[55] J. Song et al., "Demagnetization Modeling Research [64] M. Tian, H. Chen and G. Liu, "Cloud Detection and
for Permanent Magnet in PMSLM Using Extreme Learning Classification for S-NPP FSR CRIS Data Using Supervised
Machine," 2019 IEEE International Electric Machines & Machine Learning," IGARSS 2019 - 2019 IEEE International
Drives Conference (IEMDC), San Diego, CA, USA, 2019, Geoscience and Remote Sensing Symposium, Yokohama,
pp. 1757-1761, doi: 10.1109/IEMDC.2019.8785136. Japan, 2019, pp. 9827-9830, doi:
[56] I. L. Afonin, A. V. Gorelik, S. S. Muratchaev, A. S. 10.1109/IGARSS.2019.8898876.
Volkov and E. K. Morozov, "Development of an adaptive [65] E. Santi et al., "On The Use of Machine Learning
TCP algorithm based on machine learning in and Polarimetry For Estimating Soil Moisture From Radarsat
telecommunication networks," 2019 Systems of Signal Imagery Over Italian And Canadian Test Sites," IGARSS
Synchronization, Generating and Processing in 2019 - 2019 IEEE International Geoscience and Remote
Telecommunications (SYNCHROINFO), Russia, 2019, pp. Sensing Symposium, Yokohama, Japan, 2019, pp. 5760-
1-5, doi: 10.1109/SYNCHROINFO.2019.8814023. 5763, doi: 10.1109/IGARSS.2019.8900269.
[57] A. Srivastava, S. Saini and D. Gupta, "Comparison [66] D. Katsaros, G. Stavropoulos and D. Papakostas,
of Various Machine Learning Techniques and Its Uses in "Which machine learning paradigm for fake news detection?"
Different Fields," 2019 3rd International conference on 2019 IEEE/WIC/ACM International Conference on Web
Electronics, Communication and Aerospace Technology Intelligence (WI), Thessaloniki, Greece, 2019, pp. 383-387.
(ICECA), Coimbatore, India, 2019, pp. 81-86, doi: [67] S. Nakagawa, T. Hochin, H. Nomiya, H. Nakanishi
10.1109/ICECA.2019.8822068. and M. Shoji, "Prediction of Abnormal Plasma Discharge
[58] Kunal and M. Dua, "Machine Learning Approach to through Machine Learning," 2019 6th International
IDS: A Comprehensive Review," 2019 3rd International Conference on Computational Science/Intelligence and
conference on Electronics, Communication and Aerospace Applied Informatics (CSII), Honolulu, HI, USA, 2019, pp. 7-
Technology (ICECA), Coimbatore, India, 2019, pp. 117-121, 12, doi: 10.1109/CSII.2019.00009.
doi: 10.1109/ICECA.2019.8822120. [68] T. Goswami and U. B. Roy, "Predictive Model for
[59] K. Georgiadis, G. van Lankveld, K. Bahreini and W. Classification of Power System Faults using Machine
Westera, "Learning Analytics Should Analyse the Learning: Learning," TENCON 2019 - 2019 IEEE Region 10
Proposing a Generic Stealth Assessment Tool," 2019 IEEE Conference (TENCON), Kochi, India, 2019, pp. 1881-1885,
Conference on Games (CoG), London, UK, 2019, pp. 1-8, doi: 10.1109/TENCON.2019.8929264.
doi: 10.1109/CIG.2019.8847960. [69] P. Liu, W. Zhang, J. Qiu, Q. Hu and K. He, "Reagent
[60] P. P. Graybill and M. Kiani, "Eyelid Movement Color Recognition Model for Android Platform Based on
Command Classification Using Machine Learning," 2019 OPENCV and Machine Learning," 2019 International
41st Annual International Conference of the IEEE Conference on Machine Learning, Big Data and Business
Engineering in Medicine and Biology Society (EMBC), Intelligence (MLBDBI), Taiyuan, China, 2019, pp. 270-273,
Berlin, Germany, 2019, pp. 3637-3640, doi: doi: 10.1109/MLBDBI48998.2019.00060.
10.1109/EMBC.2019.8857766. [70] Y. Yuan, J. Huang and K. Yan, "Virtual Reality
[61] J. Wu, Y. Peng, M. Song, M. Cui and L. Zhang, Therapy and Machine Learning Techniques in Drug
"Link Congestion Prediction using Machine Learning for Addiction Treatment," 2019 10th International Conference
Software-Defined-Network Data Plane," 2019 International on Information Technology in Medicine and Education
Conference on Computer, Information and

www.ijcit.com 88
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

(ITME), Qingdao, China, 2019, pp. 241-245, doi: [79] D. Radovanović and S. Đukanovic, "Image-Based
10.1109/ITME.2019.00062. Plant Disease Detection: A Comparison of Deep Learning
[71] S. Ryan, R. Corizzo, I. Kiringa and N. Japkowicz, and Classical Machine Learning Algorithms," 2020 24th
"Deep Learning Versus Conventional Learning in Data International Conference on Information Technology (IT),
Streams with Concept Drifts," 2019 18th IEEE International Zabljak, Montenegro, 2020, pp. 1-4, doi:
Conference On Machine Learning And Applications 10.1109/IT48810.2020.9070664.
(ICMLA), Boca Raton, FL, USA, 2019, pp. 1306-1313, doi: [80] K. Chauhan et al., "Automated Machine Learning:
10.1109/ICMLA.2019.00213. The New Wave of Machine Learning," 2020 2nd
[72] E. Eziama, S. Ahmed, S. Ahmed, F. Awin and K. International Conference on Innovative Mechanisms for
Tepe, "Detection of Adversary Nodes in Machine-To- Industry Applications (ICIMIA), Bangalore, India, 2020, pp.
Machine Communication Using Machine Learning Based 205-212, doi: 10.1109/ICIMIA48430.2020.9074859.
Trust Model," 2019 IEEE International Symposium on [81] S. C. Lee and T. H. Han, "A 4-way Matrix Multiply
Signal Processing and Information Technology (ISSPIT), Unit for High Throughput Machine Learning Accelerator,"
Ajman, United Arab Emirates, 2019, pp. 1-6, doi: 2019 International SoC Design Conference (ISOCC), Jeju,
10.1109/ISSPIT47144.2019.9001743. Korea (South), 2019, pp. 113-114, doi:
[73] H. Lahmer, A. E. Oueslati and Z. Lachiri, "DNA 10.1109/ISOCC47750.2019.9078493.
Microarray Analysis Using Machine Learning to Recognize [82] T. Karimi, B. Majidi and M. T. Manzuri, "Deep E-
Cell Cycle Regulated Genes," 2019 International Conference School-Nurse for Personalized Health-Centered E-Learning
on Control, Automation and Diagnosis (ICCAD), Grenoble, Administration," 2019 13th Iranian and 7th National
France, 2019, pp. 1-5, doi: Conference on e-Learning and e-Teaching (ICeLeT), Tehran,
10.1109/ICCAD46983.2019.9037868. Iran, 2019, pp. 1-6, doi:
[74] N. Lisin and S. Zapechnikov, "Order-preserving 10.1109/ICELET46946.2019.9091668.
Encryption as a Tool for Privacy-Preserving Machine [83] M. A. Bhatti, R. Riaz, S. S. Rizvi, S. Shokat, F. Riaz
Learning," 2020 IEEE Conference of Russian Young and S. J. Kwon, "Outlier detection in indoor localization and
Researchers in Electrical and Electronic Engineering Internet of Things (IoT) using machine learning," in Journal
(EIConRus), St. Petersburg and Moscow, Russia, 2020, pp. of Communications and Networks, vol. 22, no. 3, pp. 236-
2090-2092, doi: 10.1109/EIConRus49466.2020.9039294. 243, June 2020, doi: 10.1109/JCN.2020.000018.
[75] S. Khatri, A. Arora and A. P. Agrawal, "Supervised [84] S. Karmakar, S. Subhasmita Swain, G. Firdous, S.
Machine Learning Algorithms for Credit Card Fraud Mohanty and T. Kumar Mohapatra, "Machine Learning
Detection: A Comparison," 2020 10th International Approach to Estimation of Internal Parameters of a Single
Conference on Cloud Computing, Data Science & Phase Transformer," 2020 International Conference for
Engineering (Confluence), Noida, India, 2020, pp. 680-683, Emerging Technology (INCET), Belgaum, India, 2020, pp.
doi: 10.1109/Confluence47617.2020.9057851. 1-4, doi: 10.1109/INCET49848.2020.9154161.
[76] M. I. Patel, S. Suthar and J. Thakar, "Survey on [85] R. Talbi, "Towards Practical Privacy-Preserving
Image Compression using Machine Learning and Deep Collaborative Machine Learning at a Scale," 2020 50th
Learning," 2019 International Conference on Intelligent Annual IEEE-IFIP International Conference on Dependable
Computing and Control Systems (ICCS), Madurai, India, Systems and Networks-Supplemental Volume (DSN-S),
2019, pp. 1103-1105, doi: Valencia, Spain, 2020, pp. 69-70, doi: 10.1109/DSN-
10.1109/ICCS45141.2019.9065473. S50200.2020.00037.
[77] U. P. Singh, K. Rakesh, Rishabh, V. Kumar and K. [86] T. Zoppi, A. Ceccarelli and A. Bondavalli, "Into the
P. Singh, "Improved Coupled Autoencoder based Zero Shot Unknown: Unsupervised Machine Learning Algorithms for
Recognition using Active Learning," 2019 IEEE Conference Anomaly-Based Intrusion Detection," 2020 50th Annual
on Information and Communication Technology, Allahabad, IEEE-IFIP International Conference on Dependable Systems
India, 2019, pp. 1-6, doi: and Networks-Supplemental Volume (DSN-S), Valencia,
10.1109/CICT48419.2019.9066127. Spain, 2020, pp. 81-81, doi: 10.1109/DSN-
[78] A. Norouzi, D. Gordon, M. Aliramezani and C. R. S50200.2020.00044.
Koch, "Machine Learning-based Diesel Engine-Out NOx [87] Y. Ma, K. Zhao, Q. Wang and Y. Tian, "Incremental
Reduction Using a plug-in PD-type Iterative Learning Cost-Sensitive Support Vector Machine with Linear-
Control," 2020 IEEE Conference on Control Technology and Exponential Loss," in IEEE Access, vol. 8, pp. 149899-
Applications (CCTA), Montreal, QC, Canada, 2020, pp. 450- 149914, 2020, doi: 10.1109/ACCESS.2020.3015954.
455, doi: 10.1109/CCTA41146.2020.9206277. [88] R. Sharma, B. Kaushik and N. Gondhi, "Character
Recognition using Machine Learning and Deep Learning - A

www.ijcit.com 89
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

Survey," 2020 International Conference on Emerging Smart Madrid, Spain, 2020, pp. 1131-1132, doi:
Computing and Informatics (ESCI), Pune, India, 2020, pp. 10.1109/COMPSAC48688.2020.0-105.
341-345, doi: 10.1109/ESCI48226.2020.9167649. [98] Z. Huang, F. Liu, M. Tang, J. Qiu and Y. Peng, "A
[89] M. Shalini and S. Radhika, "Machine Learning distributed computing framework based on lightweight
techniques for Prediction from various Breast Cancer variance reduction method to accelerate machine learning
Datasets," 2020 Sixth International Conference on Bio training on blockchain," in China Communications, vol. 17,
Signals, Images, and Instrumentation (ICBSII), Chennai, no. 9, pp. 77-89, Sept. 2020, doi: 10.23919/JCC.2020.09.007.
India, 2020, pp. 1-5, doi: [99] M. Kalimuthu, P. Vaishnavi and M. Kishore, "Crop
10.1109/ICBSII49132.2020.9167657. Prediction using Machine Learning," 2020 Third
[90] R. K. Devi and G. Elizabeth Rani, "A Comparative International Conference on Smart Systems and Inventive
Study on Handwritten Digit Recognizer using Machine Technology (ICSSIT), Tirunelveli, India, 2020, pp. 926-932,
Learning Technique," 2019 IEEE International Conference doi: 10.1109/ICSSIT48917.2020.9214190.
on Clean Energy and Energy Efficient Electronics Circuit for [100] A. Ç. Seçkin, "Adaptive Positioning System Design
Sustainable Development (INCCES), Krishnankoil, India, Using AR Markers and Machine Learning for Mobile
2019, pp. 1-5, doi: 10.1109/INCCES47820.2019.9167748. Robot," 2020 5th International Conference on Computer
[91] H. Soni, P. Arora and D. Rajeswari, "Malicious Science and Engineering (UBMK), Diyarbakir, Turkey,
Application Detection in Android using Machine Learning," 2020, pp. 160-164, doi:
2020 International Conference on Communication and Signal 10.1109/UBMK50275.2020.9219475.
Processing (ICCSP), Chennai, India, 2020, pp. 0846-0848, [101] S. Wibirama, A. P. Sidhawara, G. Lukhayu Pritalia
doi: 10.1109/ICCSP48568.2020.9182170. and T. B. Adji, "A Survey of Learning Style Detection
[92] S. Akshay and T. K. Ramesh, "Efficient Machine Method using Eye-Tracking and Machine Learning in
Learning Algorithm for Smart Irrigation," 2020 International Multimedia Learning," 2020 International Symposium on
Conference on Communication and Signal Processing Community-centric Systems (CcS), Tokyo, Japan, 2020, pp.
(ICCSP), Chennai, India, 2020, pp. 867-870, doi: 1-6, doi: 10.1109/CcS49175.2020.9231447.
10.1109/ICCSP48568.2020.9182215. [102] S. Banerji, A. Hamrick, A. Majumder, R. Menon
[93] F. O. Olowononi, D. B. Rawat and C. Liu, "Resilient and B. Sensale-Rodriguez, "Ultra-compact Design of Power
Machine Learning for Networked Cyber Physical Systems: A Splitters via Machine Learning," 2020 IEEE Photonics
Survey for Machine Learning Security to Securing Machine Conference (IPC), Vancouver, BC, Canada, 2020, pp. 1-2,
Learning for CPS," in IEEE Communications Surveys & doi: 10.1109/IPC47351.2020.9252230.
Tutorials, vol. 23, no. 1, pp. 524-552, Firstquarter 2021, doi: [103] G. Tsang, X. Xie and S. -M. Zhou, "Harnessing the
10.1109/COMST.2020.3036778. Power of Machine Learning in Dementia Informatics
[94] B. Sharma, J. K. P. S. Yadav and S. Yadav, "Predict Research: Issues, Opportunities, and Challenges," in IEEE
Crop Production in India Using Machine Learning Reviews in Biomedical Engineering, vol. 13, pp. 113-129,
Technique: A Survey," 2020 8th International Conference on 2020, doi: 10.1109/RBME.2019.2904488.
Reliability, Infocom Technologies and Optimization (Trends [104] M. Roper, "Using Machine Learning to Classify
and Future Directions) (ICRITO), Noida, India, 2020, pp. Test Outcomes," 2019 IEEE International Conference On
993-997, doi: 10.1109/ICRITO48877.2020.9197953. Artificial Intelligence Testing (AITest), Newark, CA, USA,
[95] W. Zhe, C. Wei and L. Chunlin, "DoS attack 2019, pp. 99-100, doi: 10.1109/AITest.2019.00009.
detection model of smart grid based on machine learning [105] M. Zhang and D. Wang, "Machine Learning Based
method," 2020 IEEE International Conference on Power, Alarm Analysis and Failure Forecast in Optical Networks,"
Intelligent Computing and Systems (ICPICS), Shenyang, 2019 24th OptoElectronics and Communications Conference
China, 2020, pp. 735-738, doi: (OECC) and 2019 International Conference on Photonics in
10.1109/ICPICS50287.2020.9202401. Switching and Computing (PSC), Fukuoka, Japan, 2019, pp.
[96] Y. Chang, X. Cui, G. Hou and Y. Jin, "Calibration 1-3, doi: 10.23919/PS.2019.8817991.
of the Pressure Sensor Device with the Extreme Learning [106] W. Caesarendra, T. Wijaya, B. K. Pappachan and T.
Machine," 2020 21st International Conference on Electronic Tjahjowidodo, "Adaptation to Industry 4.0 Using Machine
Packaging Technology (ICEPT), Guangzhou, China, 2020, Learning and Cloud Computing to Improve the Conventional
pp. 1-5, doi: 10.1109/ICEPT50128.2020.9202675. Method of Deburring in Aerospace Manufacturing Industry,"
[97] H. Shahriar, K. Qian and H. Zhang, "Learning 2019 12th International Conference on Information &
Environment Containerization of Machine Leaning for Communication Technology and System (ICTS), Surabaya,
Cybersecurity," 2020 IEEE 44th Annual Computers, Indonesia, 2019, pp. 120-124, doi:
Software, and Applications Conference (COMPSAC), 10.1109/ICTS.2019.8850990.

www.ijcit.com 90
 International Journal of Computer and Information Technology (ISSN: 2279 – 0764)
Volume 10 – Issue 2, March 2021

[107] S. Harish and K. S. Gayathri, "Smart Home-based Applications (CIBDA), Guiyang, China, 2020, pp. 237-240,
Prediction of Symptoms of Alzheimer’s Disease using doi: 10.1109/CIBDA50819.2020.00060.
Machine Learning and Contextual Approach," 2019 [116] J. Rahul and M. Sora, "Premature Ventricular
International Conference on Computational Intelligence in Contractions Classification using Machine Learning
Data Science (ICCIDS), Chennai, India, 2019, pp. 1-6, doi: Approach," 2020 International Conference on Smart
10.1109/ICCIDS.2019.8862163. Electronics and Communication (ICOSEC), Trichy, India,
[108] A. Yahyaoui, A. Jamil, J. Rasheed and M. Yesiltepe, 2020, pp. 367-370, doi:
"A Decision Support System for Diabetes Prediction Using 10.1109/ICOSEC49089.2020.9215290.
Machine Learning and Deep Learning Techniques," 2019 1st
International Informatics and Software Engineering
Conference (UBMYK), Ankara, Turkey, 2019, pp. 1-4, doi:
10.1109/UBMYK48245.2019.8965556.
[109] R. Karwa and V. Honmane, "Building Search
Engine Using Machine Learning Technique," 2019
International Conference on Intelligent Computing and
Control Systems (ICCS), Madurai, India, 2019, pp. 1061-
1064, doi: 10.1109/ICCS45141.2019.9065846.
[110] A. U. HAQ et al., "Identifying the Predictive
Capability of Machine Learning Classifiers for Designing
Heart Disease Detection System," 2019 16th International
Computer Conference on Wavelet Active Media Technology
and Information Processing, Chengdu, China, 2019, pp. 130-
138, doi: 10.1109/ICCWAMTIP47768.2019.9067519.
[111] H. Ouanan and E. H. Abdelwahed, "Image
processing and machine learning applications in mining
industry: Mine 4.0," 2019 International Conference on
Intelligent Systems and Advanced Computing Sciences
(ISACS), Taza, Morocco, 2019, pp. 1-5, doi:
10.1109/ISACS48493.2019.9068884.
[112] I. Ortiz Garcés, M. F. Cazares and R. O. Andrade,
"Detection of Phishing Attacks with Machine Learning
Techniques in Cognitive Security Architecture," 2019
International Conference on Computational Science and
Computational Intelligence (CSCI), Las Vegas, NV, USA,
2019, pp. 366-370, doi: 10.1109/CSCI49370.2019.00071.
[113] M. A. Rosales, A. A. Bandala, R. R. Vicerra and E.
P. Dadios, "Physiological-Based Smart Stress Detector using
Machine Learning Algorithms," 2019 IEEE 11th
International Conference on Humanoid, Nanotechnology,
Information Technology, Communication and Control,
Environment, and Management (HNICEM), Laoag,
Philippines, 2019, pp. 1-6, doi:
10.1109/HNICEM48295.2019.9073355.
[114] Q. Wu, Y. Cao, H. Wang and W. Hong, "Machine-
learning-assisted optimization and its application to antenna
designs: Opportunities and challenges," in China
Communications, vol. 17, no. 4, pp. 152-164, April 2020,
doi: 10.23919/JCC.2020.04.014.
[115] Y. Xu, "Application Research Based on Machine
Learning in Network Privacy Security," 2020 International
Conference on Computer Information and Big Data

www.ijcit.com 91