2020 8th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO)

Amity University, Noida, India. June 4-5, 2020

Big Data Analytics: Applications, Challenges &

Future Directions
Tanya Garg Surbhi Khullar
Computer Science Engineering Department Computer Science Engineering Department
Thapar Institute of Engineering & Technology Thapar Institute of Engineering & Technology
Patiala, India Patiala, India
[email protected] [email protected]

Abstract— Big data is concerned with voluminous, Online shopping platforms have been doing this in order to
complex, highly unstructured data produced from lure customers as well as increase their sales. The same case
numerous sources. It is expanding at immense rate these was seen with Starbucks, where the coffee company found
days and is a crucial issue to handle and manage the data out the pattern based on its sale patterns. Big Data is also
for the analysis of required information to save both time helpful in areas like medicine and healthcare where a vast
and cost. The data extracted can be useful for the number of medical records are analyzed to find out the
organization in various aspects. A lot of decisions have to be patterns. These patterns further help in early disease
taken by business organizations based on this massive data. detection as well as to develop new medicines. Hence, we
These decisions help in predicting future for various
have seen that Big Data is evolving at a rapid pace and we
organizations such as business, social networking sites etc.
In this paper, various aspects of big data along with big data
need to use these new torrents of data for a better and smarter
analytic techniques have been discussed in detail. The search. Big data has applications in various research areas
various big data analytic tools and techniques have also such as Machine Learning [5], Social set Analysis [6], Smart
been discussed here in this work. We also discussed various Home management using IoT [7], Business Intelligence &
challenges that are faced during the analysis of big data Decisions for customer reviews [8], Market Trends Analysis,
along with the future research areas and directions in this to build Predictive models [16], Telecommunications [17],
particular field. It is concluded that Big Data Analytics is Data Sciences etc.
emerging and highly significant field of research these days.
This Paper is organized in a systematic way so that a
Big Data analytics can have variety of applications in
different fields. The review of research carried out by
reader can easily understand the concept of Big Data
various authors have also been discussed briefly. The motive Analytics and can move towards research in this area. In
of this review paper is to help the students and researchers Section I, the concept of Big Data along with its
who knew nothing about this field but want to explore this characteristics and types has been discussed briefly. After
research area of Big Data Analytics. The reader of this work that need and Applications of Big Data Analytics have been
could solve the challenges discussed here to explore this field discussed in detail. In section II, review has been done on the
of Big Data Analytics and can smoothly move towards their research carried out by various researchers. Section III
further research. Focused on Big Data Analytic tools and techniques. In
Section IV Challenges for Big Data Analytics have been
Keywords— Big Data, Data Mining, Hadoop, Big Data discussed briefly. Section V concludes the work done in this
analytics, Machine learning, Business Intelligence paper along with the future possible directions. The aim of
writing this paper is to make people aware about the current
trends in the field of Big Data. The other objective of this
I. INTRODUCTION
paper is to focus on the challenges faced during the analysis
As we see around us, what is not digitalized? We move of Big Data and Future scope in this research area. Big Data
around with our GPS-equipped smart phones, we constantly Analysis is very important for the Business organizations for
share and exchange information and ideas over a social their development and future investments [2]. Various Big
platform e.g. Facebook, YouTube, etc. also, we perform Data Analytic Tools & Techniques have been used to pull
certain number of online transactions every day. The keynote out the suitable data. The data can exist in variety of forms. It
is, it is happening all around the globe and we are leaving a can be basically of three types: Structured, Semi-structured
digital trail with everything we do that involves a digital and Unstructured [3].
action [1]. This is happening not only at personal level, but at
a bird’s view, we see huge amounts of data is being gathered Structured Data: The data that is organized in proper
and transmitted via sensors equipped industrial machinery manner. It usually exists in the form of rows and columns
and plants. Big Data means massive data that is highly and sometimes called as labelled data. This type of data can
complex and rapidly increasing day by day due to large use either be human generated or machine generated. Only 20
of technologies. It is a challenging task to extract the relevant percent of the existing data is structured data.
information from such a complex huge data using traditional Semi-Structured Data: As its name indicates, it is
methodology. Big Data or huge volume of data is one of the neither structured nor unstructured. It is usually the data that
many driving forces behind the large number of present is complex and difficult to understand as it is not in
trends of digital transformation including artificial organized form. It is usually the data generated from logs,
intelligence, data science and the Internet of Things. emails etc. It does not generally give useful information.
Companies can predict what certain kind of customers will
want to buy, and when, with a great degree of accuracy.

978-1-7281-7016-9/20/$31.00 ©2020 IEEE 923

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.
 Unstructured Data: It is completely reverse of the x Data Generated from Mobile
structured data. It is not organized in any form. The
information extraction from this type of data is very difficult x Geographical Data
and challenging task. Big data analytic techniques are x Temporal Data
basically used to refine these types of data. 80 percent of the
data exists in this form. The machine learning techniques can x Streaming Real Time Data
also be used for analysis of this sort of data.
C. Big Data Analytics
A. Characteristics of Big Data: Big Data Analytics identifies the enormous amount of
Big Data can be categorized into five characteristics that data and discovers hidden patterns, correlations that exists
are called as 5V’s of Big Data i.e. Volume, Velocity, among those data to provide useful information. It has
Variety, Veracity, Value. become the engine for Business analytics these days.
Companies are using Big Data Analytics to identify the
future trends and growth of their business by using predictive
models of machine learning. Almost every person is
involved with use of Big Data Analytics today. All the social
networking sites and the companies like amazon and Netflix
uses Big Data Analytics to analyze the behavioral patterns
and customer interests to offer them best services at optimal
cost [19]. It has also been used in the area of scientific
research. To conclude we can say that Big Data Analytics
means processing of vast amount of data at rapid rate by the
use of tools and techniques to gain insights from the existing
data.

D. Importance of Big Data Analytics

Big Data Analytics helps the companies and
organizations to tackle and control the huge amount of data
to explore new opportunities for their business development.
Fig. 1. Characteristics of Big Data According to IIA Director, Research Tom, Big Data has
importance as discussed and show in the figure below [4]:
Volume (Data in rest): The organizations and
scientists’ study and analyze huge amount of data to gain Decrease in cost: Big Data Analytic tools helps to
insight into it. This data is collected from social media reduce large amount of data and extracts useful information
platforms, commercial transactions or sensors. which results in increase in efficiency of a system and
reduces the cost of processing that data. For Example: Cloud
Velocity (Data in motion): In the age of digitization, Analytics and Map Reduce Hadoop techniques.
sometimes the speed of data creation is more important than
the volume of data. For instance, every minute almost 72 Efficient Decision making: The techniques and tools
hours of video files are uploaded on YouTube. The need of used for Big Data Analytics helps in taking fast and accurate
the hour is how to manage this data in an effective and decisions which help business organizations to predict future
appropriate manner. trends in their business.
Variety (Data in many forms): Data is available to us Evolution of New product & Services: As Big Data
in many forms. We collect data from social media platforms, Analytic techniques helps in discovering the interests of the
online records, etc. also, this data is available to us in customers, the organizations become able to work and
different formats i.e. numeric data, text documents, generate new services for their customers which could
emails, videos, audios and transactions. impact the growth of their business significantly.
Veracity: The data available is inconsistence and
uncertain in many ways and can affect the accuracy of the
results obtained after study and analysis.

Value: The main aim of Big Data analysis is to extract

the useful information which will help us in predicting as
well as decision making.

B. Data Types
The Big Data has generated variety of data from various
sources in different domains [6]. This variety of data is to be
dealt with machine learning and Data Mining techniques to
get refined for removal of irrelevant data. The data can be
categorized into following types:
x Online Data
Fig. 2. Importance of Big Data Analytics [4]

924

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.
E. Applications of Big Data Analytics x It also helps in deceit recognition and catching tax
As discussed earlier, Big Data Analytics has applications evaders.
in various research dimensions. Some of them are discussed 4) Big Data in Media and Entertainment Industry: As
in this section below [4]:
we have realized that social media is one of the largest and
1) Big Data in Education Industry: The education most important platforms for generation of big data, we also
system has huge records of data about students, faculty, know it uses Big Data Analytics widely. Some of the ways
courses, etc. Big data Analysis can help in providing better where Big Data analysis helps are:
decisions in various sectors which will result into effective x Predicting the interests of audiences.
operation and improve performance overall. Few Areas x Optimized or on-demand scheduling of media
where Big Data has helped in bringing a change are: streams in digital media distribution platforms.
x Customized and Dynamic Learning Programs:
For effective learning, the learning programs are x Getting insights from customer reviews.
customized on the basis of student’s history. This in
x Effective targeting of the advertisements.
return also improves results of students.
5) Big Data in Transportation Industry: Big Data
x Grading Systems: Student’s data analysis has also
helped in improvising the grading systems. Analysis has helped the transportation industry to find
traffic patterns, to plan routes efficiently and to identify
x Reframing course Material: A course is designed accident prone areas. In order to identify traffic patterns
keeping in mind the only objective is the student during certain hours of the day, big data analysis has helped
should learn and gain knowledge. With the help of to manage congestion and traffic to avoid wastage of time.
student’s data one can closely observe what and how
Using the big data and predictive analysis, we can find
effectively a student is learning. Hence, data analysis
can help in modifying the course material for better accident prone areas. This helps to ensure safe travel of the
learning. masses.
6) Big Data in banking sector: The banking sector in
x Career Prediction: Proper analysis of student’s data today’s world is dynamic in nature as its data changes every
reveals the true interest, areas of weakness and second. Also, data is added and modified with every passing
strengths of the student. Therefore, one can help in millisecond. Therefore, big data analysis and predictive
predicting appropriate career choices for a student.
This analysis guides any student to find the right modeling helps in securing this data from any illegal
career for him/her. activities like money laundering, misuse of credit/debit
cards It also helps in predicting change in future in trends
2) Big Data in HealthCare: The healthcare sector is and activities.
bound to keep records of patients, doctors as well data about
different kind of diseases. This will automatically generate II. LITERATURE REVIEW
volumes of data and proper and effective analysis of this Wu et al., discussed the technique of data mining to
data has helped us in following ways: analyze the Big Data. They introduced the Big Data
x Effective analysis of data about diseases helps us to Processing model that used a mathematical Theorem called
efficiently diagnose any disease resulting into early as HACE Theorem that was based on data mining. That
treatment. model analyzed data from variety of sources and based on
HACE theorem extracted that data to get appropriate
x Proper analysis of previously available data can help information. Various challenges in the field of Big Data were
in identifying epidemics as well as controlling them also described in their research work [1].
in early stages, saving more and more lives.
Qiu et al., discussed the importance and use of Machine
x It also helps in providing evident based medicine Learning techniques for Big Data processing. The literature
which is possible due to past medical results. reviews were discussed in detail to examine the scope of
3) Big Data in Government sector: Being the machine learning techniques in the field of Big Data
Analytics. They also discussed the role and scope of Deep
government of the country, it holds records of almost
Learning, Distributed learning and representation learning to
everything. This is huge in numbers and helpful in many analyze the big data. Future Trends were also illustrated as
ways, some of which are listed below: conclusion [5].
x The records of population as a whole as well as
when divided as per some criteria helps in making R.Vatrapu et al., deliberated about the theoretical
better policies for the people. approach called as Social Set Analysis for Big Data
Analytics based on the conceptual framework. A Framework
x It helps to stay updated about natural resources like was used to combine the data from various theoretical
fossil fuels, mineral resources, agricultural land, models and after that mathematical approach was used to
forest land, livestock, etc. analyze that huge amount of data. There were some
limitations of their work that it was not applicable in the field
x It also knows proper facts and figures about
of social sciences in terms of crisis analysis, information
terrorism, unemployment other national interests and
analysis for labor rights etc. It had limited computational
does the needful according to the situation.
space that was the major drawback [6].

925

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.
 A.R. Al-Ali et al., proposed an application based on the F. Amalina et al., presented a survey on applications of
concept of Big Data Analytics. This application was Smart Big Data Analytic Techniques in solving a large complex
Energy Management System that produced energy at structured and unstructured problems using different
minimal cost. In this proposed model, each house was technologies like HDFS, Spark and map reduce. They
equipped with a data collection model based on IoT having discussed various applications of Big Data Analytics and
unique IP address forming a complex wireless mesh future trends in this field [16].
network. The data from various devices equipped in every
home was collected at the server end and there the data was According to [17], Big Data Analytics has applications in
processed. Business Intelligence concept was used to extract the area of telecommunications also. Telecommunication
useful information from the accumulated data at server [8]. industries are dealing with huge amount of data these days
due to use of mobile technology. So Big Data Analytics is
CHOI et al., described the emerging trends in Big Data very important here to process such a complex and large
Analytics that significantly impacts the Business tasks and amount of data to reduce the computational cost. They
management of risks. They discussed the challenges in the discussed the importance and scope of lambda architecture
field of Big Data Analytics for the above-mentioned used by telecommunication experts to implement Big Data
domains. The future trends were also discussed at the end Analytics in their organizations.
[9].
O. Moscoso-Zea et al., discussed the hybrid architecture
LV et al., discussed briefly the Big Data types, storage for Big data processing. Business Intelligence techniques
models, various issues related to Big data and current have been discussed that can be used for managing the
scenario of Big Data that needs to be focused. They also education system. The focus was done on examining the
discussed the security issues related to Big Data and educational data and academic procedures for the generation
provided the measures for Data Security. Finally, they of useful knowledge using variety of machine learning
introduced the various tools and technologies for managing techniques [18].
Big Data along with Future possible directions and trends
[10]. III. TOOLS FOR BIG DATA ANALYTICS
Grace P. et al., described the importance of Big Data Big Data Analytics is useful in areas where better
Analytics in the field of Business Intelligence. They mainly decision making as well as predictive analysis matter. Our
focused on the Cost-effective Data Processing techniques main aim is to study and analyze the data in an effective and
that helps in taking business decisions instantly. This efficient manner. There are certain data software available
research was totally business oriented. A business-oriented known as big data tools which help in productive analysis of
model was proposed for Business Decisions called as IRIS the data. Some of the tools are [23], [16]:
based on Spark [11].
1) Apache Hadoop: Apache Hadoop is an open source
J.Zhao et al., elaborated the concept of sustainable framework, used for clustered file system and managing
architecture for Big Data Analytics. They discussed the
concept of Data Analytic systems that have the capability of big data by means of MapReduce programming model.
analyzing online data. They analyzed the complex nature of The main features of Hadoop that made it widely used
DAS and based on that they proposed a framework for the among masses like Amazon are: The HDFS or Hadoop
development of architecture for DASs. They designed an Distributed File System has the ability to hold all type of
intelligent system for Data Analytics [12]. data. It can be audio, video, images, plain text and XML
Rocha Silva discussed the importance of Analytical also. Hadoop is extensible and provides rapid data access.
Intelligence associated with the technology of Data Sciences. Hadoop is highly useful in R&D sector. At the same time
A lot of research papers have been studied by him to describe
Hadoop uses more disk space due to its thrice data
the importance of Analytical Intelligence [13].
redundancy.
Sujata & Subash K.S. discussed the machine learning 2) Cloudera Distribution for Hadoop: Cloudera
approach for analysis of Big Data, an ensembled
classification algorithm was used to analyze the customer Hadoop Distribution provides extensible and flexible
reviews and to draw results from it. Text Analytics approach platform. Because of this Cloudera results in better
has been used for the analysis of the customer reviews. The performance while managing larger volume and more of
predictive model based on machine learning approach variety data. Cloudera provides more potential to process
predicted results with the accuracy of 79.83%. The
and analyze unlimited data. This makes it popular among
performance was compared with other classifiers also. The
proposed approach could be used to analyze the Customer major companies. There are automated wizards in Cloudera
interest which can help in taking business decisions that help to deploy the cluster faster. This makes it easy to
accurately [14]. manage Hadoop. Cloudera is integrated in its nature
According to ALAKA et al., Big Data Analytical therefore it provides highly secure environment for data
approach can be used to design various predictive models analysis.
based on which numerous decisions can be taken. A Big 3) Cassandra: Apache Cassandra is an open source,
Data Analytic model called as FPM (Failure Predictive distributed NoSQL database management system. It is used
Model) was proposed that used the Artificial Neural
Networks Algorithm to save time required for the to manage large volume of data. It also uses CQL i.e.
computational problems [15]. Cassandra Structure Language to interact with database.
Companies like American Express, Facebook, Yahoo use

926

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.
 Cassandra. Cassandra is highly available as it can manually. There are several advantages of Big Data Analytic
efficiently manage data that is spread across a number of Techniques despite it poses various challenges also which is
a serious issue. Some of the issues are related to data security
servers. As it is distributed in nature there is no single point
and privacy, complex coding, difficulty in implementation
failure. It has log -structured storage and linear scalability. etc. Researchers should focus more on providing best
4) KNIME: KNIME stands for Konstanz Information optimal solutions to solve those challenges. The challenges
Miner. It is an open source tool which is widely used for can be broadly categorized into three types that are discussed
research, data mining, data analytics, and Business below [20]:
Intelligence. It is sometimes considered as a good choice 1) Data Challenges: The main challenge is to handle
because it integrates easily with other technologies and Big Data storage, processing and to discover hidden patterns
languages. It also supports Linux, OS X, and Windows among those data. The main data challenges that needs to be
Operating Systems. Companies like Comcast, Johnson & focused are:
Johnson etc. Cassandra has an abundant algorithm set for x Data Storage
efficient computations. It automates manual work x Velocity of Data
productively hence reduces workload. KNIME is easy to
x Data Variety
set up and has ETL operations.
5) MongoDB: MongoDB is a document-oriented x Computational power of data
database. It is written in C, C++ and JavaScript. It supports x Data Understanding
multiple operating systems. Companies like MetLife,
Google, Facebook uses MongoDB. MongoDB allows x Quality of Data
Aggregation, Indexing, Replication, Server–side execution x Data Representation
of java script and file storage. It is easy to learn, install and
maintain. Also, reliable and low-cost tool. Sometimes 2) Management Challenges: Now a days many
MongoDB produces limited analyses and at slow pace as Business industries are equipped with Big data
compared to other tools. Technologies. So, it is very important for them to choose
6) Spark: It is also an open source framework for data best Big Data Analytics techniques and to get maximum
processing that uses Python, Java, R algorithms. Spark is benefit from them. The company has to manage the changes
100 times faster than Hadoop. timely. Some of the challenges are:
7) Rapid Miner: It is a tool used for Data mining task x Leadership
x Talent Management
and it is also freely available. It is based on GUI that is
x Management of Technology
used for predictive analysis. This tool was developed by x Decision taking capability
Rapid Miner studio. It is basically used to analyze semi- x Environment of company
structured type of data. 3) Process Challenges: The experts can analyze the
8) Weka: Weka stands for Waikato Environment for model completely on the basis of past experiences but still
Knowledge Analysis. It is a data mining tool developed by they lack in some aspects which is a challenge in the task of
University of Waikato, New Zealand. This tool consists of processing Big data. Selection of right model for the data
classification algorithms that evaluates the performance of analysis is a challenging task that needs to be considered
seriously. The process challenges are as following:
machine learning models.
x To capture the Data from various sources
9) Talend: This tool performs the task using Hadoop
Technology. It is also an open source tool but due to its x Conversion of collected data into analyzable form
many drawbacks, it is not used these days. x Discovering the hidden patterns among the data
10) NodeXL: This tool is available in both modes:
x Analyzing the scope of techniques required to filter
Free as well as paid. It depends upon usage that means the data.
whether it is to be used for individual purpose or
commercial. V. CONCLUSIONS AND FUTURE SCOPE
11) Orange: It is also a data mining and machine
In this paper, we discussed the concept of Big Data
learning tool. It is mainly used for the purpose of Analytics in detail. Initially in Section I, Big Data has been
visualization. It is also freely available. It consists of a discussed in detail along with its characteristics. After that
toolbox having variety of algorithms for the analysis of the concept of Big Data Analytics has been discussed. The
data. applications and importance of Big Data Analytics have been
described in detail. We focused on the current scenario of
IV. CHALLENGES IN BIG DATA ANALYTICS Big Data Analytic Tools also. After that challenges that are
creating obstacle on the way of Big Data Analytics have
Big Data generally means huge amount of data that is been summarized. The recent work that has been done by
very difficult to process and store. Big Data Analytics is used various researchers along with their benefits and limitations
to analyze Big Data to convert it into processed and relevant have also been discussed. The role of Machine Learning
form by removing redundancy and irrelevant data by using Techniques in the field of Big Data has also been discussed.
various kinds of techniques. Big Data cannot be processed To conclude we can say that Big Data Analytics have both

927

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.
positive and negative impacts. Big Data Analytics is the [9] T. M. Choi, H. K. Chan, and X. Yue, “Recent Development in Big
latest emerging research area in the field of Big Data Data Analytics for Business Operations and Risk Management,”
IEEE Trans. Cybern., vol. 47, no. 1, pp. 81–92, 2017, doi:
Processing. Big Data Analytics is beneficial for various 10.1109/TCYB.2015.2507599.
fields like in Education, Healthcare, Agriculture, Military, [10] Z. Lv, H. Song, P. Basanta-Val, A. Steed, and M. Jo, “Next-
Banking, especially for Business Organizations. Generation Big Data Analytics: State of the Art, Challenges, and
Future Research Topics,” IEEE Trans. Ind. Informatics, vol. 13, no. 4,
In Future, one can work on finding the solution for the pp. 1891–1899, 2017, doi: 10.1109/TII.2017.2650204.
various challenges that are creating problem on the path of [11] G. Park, L. Chung, L. Zhao, and S. Supakkul, “A goal-oriented big
Big Data Analytics. Further the work can be done on data analytics framework for aligning with business,” Proc. - 3rd
designing new techniques and algorithms for the processing IEEE Int. Conf. Big Data Comput. Serv. Appl. BigDataService 2017,
of Big data. The collection of data from variety of sources pp. 31–40, 2017, doi: 10.1109/BigDataService.2017.29.
can be improved by developing different new techniques. [12] J. Zhao, Y. Liu, and P. Zhou, “Framing a Sustainable Architecture for
More focus should be given to the data security issues which Data Analytics Systems: An Exploratory Study,” IEEE Access, vol. 6,
pp. 61600–61613, 2018, doi: 10.1109/ACCESS.2018.2872673.
have not considered in this work but it is very crucial issue
[13] F. A. R. Silva, “Analytical intelligence in processes: Data science for
that need to be discussed. We have not provided the solutions business,” IEEE Lat. Am. Trans., vol. 16, no. 8, pp. 2240–2247, 2018,
to face challenges that exists in the field of Big Data doi: 10.1109/TLA.2018.8528241.
Analytics. Researchers can continue their future work for [14] S. Atul Khedkar and S. K. Shinde, “Customer Review Analytics for
solving the challenges that are discussed here in this work. Business Intelligence,” 2018 IEEE Int. Conf. Comput. Intell. Comput.
The techniques should be developed for the analysis of real Res. ICCIC 2018, pp. 1–5, 2018, doi: 10.1109/ICCIC.2018.8782305.
time data instead of examining the static past data. [15] H. Alaka, L. Oyedele, H. Owolabi, O. Akinade, M. Bilal, and S.
Ajayi, “A Big Data Analytics Approach for Construction Firms
REFERENCES Failure Prediction Models,” IEEE Trans. Eng. Manag., vol. 66, no. 4,
pp. 689–698, 2019, doi: 10.1109/TEM.2018.2856376.
[1] X. Wu, X. Zhu, G. Q. Wu, and W. Ding, “Data mining with big data,”
IEEE Trans. Knowl. Data Eng., vol. 26, no. 1, pp. 97–107, 2014, doi: [16] F. Amalina et al., “Blending Big Data Analytics: Review on
10.1109/TKDE.2013.109. Challenges and a Recent Study,” IEEE Access, vol. 8, pp. 1–1, 2019,
doi: 10.1109/access.2019.2923270.
[2] “Introduction to Big Data,” Accessed on: Jan.
29,2020..[Online]Avaialble:https://www.sas.com/en_in/insights/big- [17] H. Zahid, T. Mahmood, A. Morshed, and T. Sellis, “Big data
data/what-is-big-data.html analytics in telecommunications: literature review and architecture
recommendations,” IEEE/CAA J. Autom. Sin., vol. 7, no. 1, pp. 18–
[3] “Types of Big Data,” Knowledgehut Blog, Last Updated on Oct 38, 2019, doi: 10.1109/JAS.2019.1911795.
3,2019. Accessed on Jan 29, 2020. [Online] Available
:https://www.knowledgehut.com/blog/big-data/types-of-big-data [18] O. Moscoso-Zea, J. Castro, J. Paredes-Gualtor, and S. Lujan-Mora,
“A Hybrid Infrastructure of Enterprise Architecture and Business
[4] “Introduction to Big data Analytics,” Accessed on Jan 29,2020. Intelligence Analytics for Knowledge Management in Education,”
Availabe:https://www.sas.com/en_in/insights/analytics/big-data- IEEE Access, vol. 7, pp. 38778–38788, 2019, doi:
analytics.html 10.1109/ACCESS.2019.2906343.
[5] Qiu, Q. Wu, G. Ding, Y. Xu, and S. Feng, “A survey of machine [19] K.A. Somani et al., Big Data Analytics, Tools and Technology for
learning for big data processing,” EURASIP J. Adv. Signal Process., Effective Planning, London New York: CRC Press, Dec 2017.
vol. 2016, no. 1, 2016, doi: 10.1186/s13634-016-0355-x.
[20] “Applications of Big data,” Accessed on Jan 29,2020. [Online]
[6] R. Vatrapu, R. R. Mukkamala, A. Hussain, and B. Flesch, “Social Set Available: https://intellipaat.com/blog/7-big-data-examples-
Analysis: A Set Theoretical Approach to Big Data Analytics,” IEEE application-of-big-data-in-real-life/
Access, vol. 4, pp. 2542–2571, 2016, doi:
10.1109/ACCESS.2016.2559584. [21] “Supervised and Unsupervised Learning,” Accessed on Jan 29,2020.
[Online]Available:https://www.geeksforgeeks.org/supervised-
[7] A. Praveena and B. Bharathi, “A survey paper on big data unsupervised-learning/
analytics,”2017 Int. Conf. Inf. Commun. Embed. Syst. ICICES 2017,
no. Icices, 2017, doi: 10.1109/ICICES.2017.8070723. [22] S. L. Nita, “Machine Learning Techniques Used in Big Data,” Sci.
Bull. Nav. Acad., vol. 19, no. 1, pp. 466–471, 2016, doi:
[8] A. R. Al-Ali, I. A. Zualkernan, M. Rashid, R. Gupta, and M. Alikarar, 10.21279/1454-864x-16-i1-078.
“A smart home energy management system using IoT and big data
analytics approach,” IEEE Trans. Consum. Electron., vol. 63, no. 4, [23] “Big Data Tools,” Accessed on Jan 29,2020.[Online] Available
pp. 426–434, 2017, doi: 10.1109/TCE.2017.015014. :https://www.softwaretestinghelp.com/big-data-tools/

928

Authorized licensed use limited to: UNIVERSITY OF ROCHESTER. Downloaded on September 21,2020 at 03:58:04 UTC from IEEE Xplore. Restrictions apply.