Content uploaded by Preeti Gulia
Author content
All content in this area was uploaded by Preeti Gulia on Oct 04, 2022
Content may be subject to copyright.
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075 (Online), Volume-8 Issue-12, October 2019
4910
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L35501081219/2019©BEIESP
DOI: 10.35940/ijitee.L3550.1081219
Journal Website: www.ijitee.org
Abstract: Now-a-days artificial intelligence has become an
asset for engineering and experimental studies, just like statistics
and calculus. Data science is a growing field for researchers and
artificial intelligence, machine learning and deep learning are
roots of it. This paper describes the relation between these roots of
data science. There is a need of machine learning if any kind of
analysis is to be performed. This study describes machine learning
from the scratch. It also focuses on Deep Learning. Deep learning
can also be known as new trend of machine learning. This paper
gives a light on basic architecture of Deep learning. A
comparative study of machine learning and deep learning is also
given in the paper and allows researcher to have a broad view on
these techniques so that they can understand which one will be
preferable solution for a particular problem.
Keywords : Machine Learning, Deep learning, Artificial
Intelligence, shallow learning.
I. INTRODUCTION
In the era of data sciences, artificial intelligence is trying
to provide human kind intelligence to the computer and for
this machine learning and deep learning are the technologies
which are helping artificial intelligence to do it. Machine
Learning is the branch or subset of artificial intelligence that
train the machines how to learn. Deep learning is confined
version of machine learning. It helps to raise the high
standards of learning environment. Machine learning and
deep learning both plays vital role in upgrading the computer
systems to be an expert systems that can take decisions and
make predictions without a human intervention.
Artificial intelligence is a field which helps computer
system to be intelligent and take decisions. Machine learning
helps to implement Artificial Intelligence on the system and
deep learning helps to achieve machine learning goals on the
system more systematically. Figure 1 shows it pictorially.
This paper is divided into two parts. In section II, will
explain machine learning, its procedures, its applications etc.
In section III, Different approaches of machine learning are
discussed such as deep learning and shallow learning.
Revised Manuscript Received on October 30, 2019.
* Correspondence Author
Ayushi Chahal*, Department of Computer Science and Applications,
Maharishi Dayanand University, Rohtak, India. Email:
ayushichahal@gmail.com
Preeti Gulia, Department of Computer Science and Applications,
Maharishi Dayanand University, Rohtak, India. Email:
research.mdu81@gmail.com
© The Authors. Published by Blue Eyes Intelligence Engineering and
Sciences Publication (BEIESP). This is an open access article under the
CC-BY-NC-ND license http://creativecommons.org/licenses/by-nc-nd/4.0/
It describes each one’s different methods and different
algorithms used by them. In section IV, a comparative study
between deep learning and other conventional methods of
machine learning.
Fig. 1. AI, machine learning and deep learning
paradigm
II. MACHINE LEARNING
Machine learning is based on the idea that system can learn
from data, identify the patterns and make decision with
minimum human intervention [2]. This is the scientific study
of algorithms and statistical models with the help of which
computer system perform a specific task without using
instruction, inference and patterns. Machine learning
algorithms build mathematical model based on sample data
and then make the decision.
A. Machine learning procedure
Machine learning incorporates four steps, given below
(shown in the figure 2):
- First, feature extraction
- Second, selection of corresponding machine learning
algorithm
- Third, training and evaluation the data model’s efficiency
- Four, using trained model for prediction
B. Requirements to Create Good Machine Learning
Systems:
Data preparation capabilities
Basic and Advanced algorithms
Scalability
Various processes i.e. Automation and Iterative
Ensemble modeling
Machine Learning and Deep Learning
Ayushi Chahal, Preeti Gulia
Machine Learning and Deep Learning
4911
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L35501081219/2019©BEIESP
DOI: 10.35940/ijitee.L3550.1081219
Journal Website: www.ijitee.org
Fig. 2. Machine learning model
C. Relationship with Other Fields:
Machine learning is considered as the subset of artificial
intelligence. In earlier days of AI as academic discipline,
researchers were interested in having machine learn. They
attempted to solve the problem with various symbolic
methods as well as connectionist approach where neural
network, pattern recognition are used. In 1990s, Machine
learning is reorganized as a separate field. It shifted focus
from symbolic approach to the methods and models of
statistics and probability theory [4].
Relation to data mining: Both of these employ same
methods often and overlap with each other. But machine
learning focuses on prediction based on known properties
while data mining focuses on the discovery of unknown
properties. Data mining uses machine learning methods,
machine learning also employs data mining methods; but with
different goals or to improve the learner accuracy.
Relation to optimization: Machine learning is also
intimated with optimization. Learning problems are
formulated as minimization of loss function. Loss functions
show the discrepancy between prediction of model and actual
problem.
Relation to statistics: It is also closely related with
statistics. The ideas of machine learning have had a
relationship with statistics from methodological principles to
theoretical tools such as the modeling paradigm.
D. Who’s Using Machine Learning?
As the industries grow, large volumes of data have been
recognized. For handling that data, machine learning
technology is required. With the machine learning,
organizations are able to work more efficiently. Machine
learning is used in following areas:
Financial services: In financial services, machine
learning technology is used to identify the important insight in
data and to prevent fraud. The insights help to identify
investment opportunities or help investors to know when to
trade. Data mining concepts also identify high risk profiles of
clients or to pinpoint warning signs of fraud.
Health Care: This is the major area in which wearable
devices and sensors are used to assess patient’s health in real
time. Machine learning also helps medical experts to analyze
the data to identify trends. This may lead to improve
diagnoses and treatment.
Government sector: Government agencies use machine
learning to mine the data for insight where agencies like
public safety and utilities etc. have multiple sources of data.
Sensor data analysis increases the efficiency and save money.
Machine learning can also be used for security purpose i.e.
help to detect fraud and to minimize the identity theft.
Retail sector: In retail sector, machine learning is used
to analyze the buying history of customers. Retailers rely on
machine learning to capture data, analyze and use it to
personalize the shopping experience. It is also helpful to
implement the marketing campaign, optimizing price, and for
customer insights.
Transportation: Machine learning is used to make
routes more efficient and to predict the problems to increase
profitability. It can be done after analyzing the data to identify
patterns and trends. Data analysis and modeling aspects are
key factors to delivery companies and transportation
organizations.
Oil and gas: In this sector, machine learning is used to
find new energy source and to analyze minerals in ground. It
is also used to predict refinery sensor failure. Streamlining
oil distribution makes it more efficient and economic.
E. Processes and Techniques associated with
machine learning:
A number of processes, techniques and methods can be
applied to enhance the performance of machine learning and
these are as follows:
Feature learning
Sparse dictionary learning
Anomaly detection
Decision tree
Association rules
F. Applications of Machine learning:
There are many applications of machine learning such as:
Adaptive websites
Bioinformatics
Brain-machine interface
Computer vision
Data quality
DNA sequence classification handwriting recognition
Machine learning control
User behavior analytics etc…
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075 (Online), Volume-8 Issue-12, October 2019
4912
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L35501081219/2019©BEIESP
DOI: 10.35940/ijitee.L3550.1081219
Journal Website: www.ijitee.org
III. MACHINE LEARNING APPROACHES
Basically, Machine learning methods are broadly
categorized in two categories i.e. Shallow learning and deep
learning [16]. Shallow learning basically uses neural networks
with single layers or SVMs (Support Vector Machines) while
deep learning uses neural network with more than one hidden
layers. As shown in figure 3:
Fig. 3. Different approach of machine learning
A. Shallow Learning
Shallow learning is broadly divided into two categories:
Supervised and Unsupervised Learning. But there are also
other methods of machine learning. Overview of popular
methods is as follows:
Supervised learning: In supervised learning,
algorithm builds a mathematical model from a set of data
that contains both the input and desired outputs.(wiki)
These algorithms are trained using labeled examples i.e.
input and desired outputs are known. In this learning,
algorithm receives a set of inputs along with corresponding
correct outputs. Algorithm learns by comparing its actual
output with correct outputs to find out errors. Then, model
is modified accordingly. Classification, regression,
prediction and gradient boosting are the example of
supervised learning which use pattern to predict the values.
This learning is commonly used in those applications where
historical data predicts future events. Classification and
regression are the tasks that are performed by supervised
learning. Some examples of supervised machine learning
are Nearest neighbor, Naïve Bayes, Decision Tree,
Regression Tree etc. Figure 4 gives the pictorial view of
different method of supervised learning.
Unsupervised learning: In unsupervised learning, a
mathematical model is to be built from a set of data which
contains only inputs. Desired output labels are not present
in this type of learning. Unsupervised learning is used
against that data which doesn’t consists historical label.
K-means, Association Rules are an example of such
algorithms. Figure 5 describes different methods of
unsupervised learning.
Fig. 4. Supervised Learning
Semi-supervised learning: In some cases, input may
be only partially available, or restricted to special feedback.
At that time, these algorithms are used. These are used to
develop mathematical model from incomplete training data,
where a portion of the sample input doesn’t consist
labels. This learning is useful when cost of labeling is too
high to allow for fully labeled training process.
Reinforcement learning: This is the area of learning
concerned with how software agents take actions in an
environment to maximize the cumulative reward. In this
type of learning, a feedback is to be given in the form of
positive or negative reinforcement in a dynamic
environment. These are commonly used in autonomous
vehicle or in learning to play game against human opponent
[3]. Q-learning is an example of reinforcement learning.
Active learning: Desired outputs are accessed for a
limited set of inputs. In this learning, the inputs are based on
budget, and optimize the choice of inputs for which output
will be acquired.
Meta learning: Here, algorithms learn their own
inductive bias based on previous experiences. Some
examples of meta learning are Bagging, Boosting, Random
Forest.
Fig. 5. Unsupervised Learning
B. Deep Learning
Deep learning is a set of algorithms of machine learning
which uses multiple layers that corresponds to different level
of abstraction to each level. . It consists of input layer, output
layer and several hidden layer. It is used for voice synthesis,
image processing, handwriting recognition, object detection,
prediction analytics and decision making. [10] Deep learning
can be broadly classified into three types (figure 6):
Fig. 6. Deep learning
Machine Learning and Deep Learning
4913
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L35501081219/2019©BEIESP
DOI: 10.35940/ijitee.L3550.1081219
Journal Website: www.ijitee.org
Generative models: Generative models are used for
unsupervised learning. It includes algorithms like Deep
Belief Network (DBN), Deep auto-encoders, Deep
Boltzmann (DBM).
Discriminative models: Discriminative models usually
provide supervised learning approaches. It involves
Convolution Neural Network (CNN), Deep Stacking
Network (DSN).
Hybrid models: Hybrid models incorporate the benefits of
both discriminative and generative models. Deep Neural
network (DNN) is an example of hybrid models.
C. Deep Learning architecture
Deep Learning consists of supervised or unsupervised learning
techniques based on many layers of artificial neural networks
that are able to learn hierarchical representations in deep
architectures. [11] It is extended version of artificial neural
network. Deep Learning architectures consist of multiple
processing layers. Each layer is able to produce non-linear
responses based on the data from its input layer.
The functionality of Deep Learning is imitated from the
mechanisms of human brain and neurons for processing of
Fig. 8. Deep Neural network
signals. Deep Learning architectures have gained more
attention in recent years compared to the other traditional
machine learning approaches. Such approaches are considered
as being shallow-structured learning architectures versions
(i.e., a limited subset) of Deep Learning.
A Deep Neural Network consists of an input layer,
severalhidden layers, and an output layer. Each layer includes
severalunits called neurons. These neurons are also called as
artificial neurons. A neuron receives several inputs, performs
a weighted summation over its inputs, then the resulting sum
goes through an activation function to produce an output.
Each neuron has a vector of weights associated to its input
size as well as a bias that should be optimized during the
training process. Figure 7 below shows the structure of
neuron.
Fig. 7. Structure of a neuron
When these artificial neurons are assigned sequentially which
makes a chain as one neurons output becomes input of next
neuron, and this process goes on over and over which makes a
Artificial Neural Network. Deep Learning Neural Networks
consists of more than one hidden layer as shown below in
figure 8.
IV. DEEP LEARNING COMPARISON WITH
CONVENTIONAL MACHINE LEARNING
TECHNIQUES
Deep learning is a new era of machine learning. Deep
learning includes both supervised and unsupervised learning
paradigm of machine learning. Machine learning and deep
learning helps in providing intelligence to the system that can
make prediction for future using past data.[12]
Conventional machine learning algorithms can’t learn
directly from the raw data. They need careful engineering
to carefully extract features from raw data and highly
classified domain expertise, which are further used to in
internal representations to identify these feature’s
patterns. In Deep Learning, first step of machine learning
procedure is not present. This step is automated in deep
learning. Deep Learning can extract new features
automatically from raw data. Figure 9 shows this point
clearly [13].
Fig. 9. Feature extraction is automated in deep
learning
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075 (Online), Volume-8 Issue-12, October 2019
4914
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: L35501081219/2019©BEIESP
DOI: 10.35940/ijitee.L3550.1081219
Journal Website: www.ijitee.org
Deep learning algorithms work more accurately on large
Data set as compared to conventional machine learning
algorithms. While machine learning algorithms
outperforms deep learning in case of small or medium size
datasets. [14]
Deep learning algorithms take less time to infer a problem
as compared to conventional machine learning algorithms
Deep learning performs a high amount of matrix multiple
hence it needs powerful engine preferably GPU (Graphical
Processing Units) or specially designed TPU (Tensor
Processing Units) while other conventional machine
learning algorithms can work on low end machines.
Deep learning algorithms are difficult to impossible to
interpret. Some of the machine learning algorithms like
(logistics, decision tree) can be interpreted easily while
some (like SVM) are almost impossible to interpret. [15]
Training time for data to create the model is more in deep
learning as compared to other machine learning algorithms.
V. CONCLUSION
This article examined the concepts of machine learning.
Machine learning has gained a lot of attention of researchers
nowadays due to its distinct features. Firstly, the article
specified the points to make a good machine learning system.
Followed by this, the usage and applications of machine
learning have been discussed in this article. However the road
of machine learning is not as simple as it looks to be. There
are some challenges in this area to get the expected results
such as lack of suitable data, data bias, and lack of resources,
privacy problems and evaluation problems. This paper crates
a broad view for a researcher for machine learning by
categorizing it into two parts, namely: shallow learning and
deep learning. Supervised and unsupervised machine learning
concepts are supposed to be in the category of shallow
learning as these techniques use less number of hidden layers
or SVMs. While deep learning is considered as a different
category, because of its deep layered architecture discussed in
the article.
Deep learning is a growing field in a sector of predictive
analytics. This paper provides a comparative study of
conventional methods of machine learning and deep learning
which helps new researchers to choose which technique
would be right to apply in a particular environment. Such as, if
one is working on small training data set then he must use
machine learning algorithms rather than deep learning while,
if dataset needed to choose the features then one must use
machine learning technique because in case of deep learning
this feature selection procedure is automated researcher do
not have to bother about it. This paper creates base for the
researcher who wants to pursue research in field of artificial
intelligence or predictive analytics.
REFERENCES
1. https://en.wikipedia.org/wiki/Machine_learning
2. https://www.sas.com › SAS Insights › Analytics Insights
3. Bishop, C. M. (2006), Pattern Recognition and Machine Learning,
Springer, ISBN 978-0-387-31073-2
4. 11. Langley, Pat (2011). "The changing science of machine
learning". Machine Learning. 82 (3): 275–
279. Doi:10.1007/s10994-011-5242-y
5. Alpaydin, Ethem (2010). Introduction to Machine Learning. MIT
Press. p. 9. ISBN 978-0-262-01243-0
6. Why Machine Learning Models Often Fail to Learn: QuickTake
Q&A". Bloomberg.com. 2016-11-10. Retrieved 2017-04-10.
7. "The First Wave of Corporate AI Is Doomed to Fail". Harvard Business
Review. 2017-04-18. Retrieved 2018-08-20.
8. "Why the A.I. euphoria is doomed to fail". VentureBeat. 2016-09-18.
Retrieved 2018-08-20.
9. "9 Reasons why your machine learning project will
fail". www.kdnuggets.com. Retrieved 2018-08-20.
10. P. Kuang, W.-N. Cao, and Q. Wu, “Preview on structures and
algorithms of deep learning,” in 2014 11th International Computer
Conference on Wavelet Actiev Media Technology and Information
Processing(ICCWAMTIP), Chengdu, China, 2014, pp. 176–179.
11. L. Deng, “Deep Learning: Methods and Applications,” Found.
Trends® Signal Process., vol. 7, no. 3–4, pp. 197–387, 2014.
12. J. L. Berral-García, “When and How to Apply Statistics, Machine
Learning and Deep Learning Techniques,” in 2018 20th International
Conference on Transparent Optical Networks (ICTON), 2018, pp.
1–4.
13. J. Latif, C. Xiao, A. Imran, and S. Tu, “Medical Imaging using
Machine Learning and Deep Learning Algorithms: A Review,” in 2019
2nd International Conference on Computing, Mathematics and
Engineering Technologies (iCoMET), 2019, pp. 1–5.
14. N. G. Paterakis, E. Mocanu, M. Gibescu, B. Stappers, and W. van Alst,
“Deep learning versus traditional machine learning methods for
aggregated energy demand prediction,” in 2017 IEEE PES Innovative
Smart Grid Technologies Conference Europe (ISGT-Europe), 2017,
pp. 1–6.
15. M. Sewak, S. K. Sahay, and H. Rathore, “Comparison of Deep
Learning and the Classical Machine Learning Algorithm for the
Malware Detection,” in 2018 19th IEEE/ACIS International
Conference on Software Engineering, Artificial Intelligence,
Networking and Parallel/Distributed Computing (SNPD), 2018, pp.
293–296.
16. A. Jabeen, N. Ahmad, and K. Raza, “Machine Learning-based
state-of-the-art methods for the classification of RNA-Seq data,”
Bioinformatics, preprint, Mar. 2017.
AUTHORS PROFILE
Ms. Ayushi Chahal has completed her M.Tech
from GJUS&T University. She is currently
pursuing Ph.D. in Computer Science at Department
of Computer Science & Applications,
M.D.University, Rohtak. Her main research area
includes network security, Internet of Things (IoT),
Machine Learning.
Dr. Preeti Gulia is currently working as
Assistant Professor at Department of Computer
Science & Applications, M.D.University, Rohtak,
India. She is serving the Department since 2009.
She earned her doctoral degree in 2013. She has
published more than 65 research papers and articles
in journal and conferences of National/
International repute including ACM, Scopus. Her
area of research includes Data Mining, Big Data,
Machine Learning, Deep Learning, IoT, Software Engineering. She is an
active professional member of IAENG, CSI and ACM. She is also serving as
Editorial Board Member Active Reviewer of International/ National
Journals. She has guided one research scholar as well as guiding four Ph.D.
research scholars from various research areas..