International Journal of Science and Research (IJSR)

ISSN (Online): 2319-7064

Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438

A Review Paper on Big Data Analytics

Ankita S. Tiwarkhede1, Prof. Vinit Kakde2
1
ME (CSE) 2ND Sem, GHRCEM, SGBAU Amravati University, Amravati, India
2
GHRCEM, Amravati, India, SGBAU Amravati University

Abstract: We live in on-demand world with vast majority of data. People and devices are constantly generating data, while streaming a
video, active in social media, playing games, search any location using GPS. This data increase day by day from many resources,
various types of techniques and technologies. The data is categories as “Big Data”. Big Data is huge in Variety, Velocity and Sheer
volume. It is structured and unstructured data and heterogeneous in nature. The goal of Big Data analysis is to extract useful values,
suggest conclusions and/or support decision making. In this topic, we provide an extensive survey of big data analytics research, while
highlighting the specific concern in big data world. According to Application evolution, we discuss six types of big data application such
as structured data analytics, Text analytics, Web analytics, Multimedia analytics, and Mobile analytics. We illustrate the techniques of
analyzing the big data such as A/B testing, classification, crowdsourcing, and data mining.

Keywords: Big data management, Big data, Analytics, Analyzing Technique

1. Introduction i.e. Unknown correlations, market trends, customer

preferences and other useful information’s [16]. The
Big Data term appeared for First time in 1998 in Silicon analytics can more lead to more effective marketing, better
Graphics (SGI) Slide Deck By john Massey with the title of customer services.
Big Data [3]. Big Data is very vast in majority and Complex
data. Heterogeneity, scale, timeliness, complexity, and Big data analytics project are rapidly emerging as the
privacy problems with big data hamper the progress at all preferred solution to address business and technology trends
phases of the process that can create value from data [5]. that are disrupting traditional data management processing
There are various resources of Big Data For Example: Audio, [10]. Analytics helps to discover what has changed and the
Videos, and Post in Social Media, Various Database Tables, possible solutions [5].With big data analytics, the user is
and Email Attachment etc. People uses twitter in diverse trying to discover new business facts that no one in the
form and store 250 Million tweets Per Day. 4 Billion People
watching YouTube per Day. Nowadays, Data produced in Enterprise knew before [7]. We introduce literature survey of
Zettabytes. Big data has many opportunities like financial big data analytics in section 2. Section 3 contains background
services, Healthcare, Retail, Web/social, Manufacturing and and overview of big data. Section 4 contains big data
Government [10]. Big data has now reached every sector in analytics in detail and section 5 concludes the paper.
the global economy. We estimate that by 2005, nearly all
sectors in the US economy had an average of 200 terabytes 2. Literature survey
of stored data per company with more than 1,000 employees
[12]. Big data moving continue to evolve rapidly, driven by Over the last many years, there are many researchers has
innovation in underlying technologies. In August 2010, The completed their work successfully on big data. Hundreds of
White house, OMB, Proclaimed that big data is national articles have appeared in the general business press (For
challenge and priority along with healthcare and national example Forbes, Fortune, Bloomberg, Business week, The
security [14]. Wall street journal, The Economist)[1]. National Institute of
Standards and Technology [NIST] said that Big Data in
Traditional data management and analysis system mainly which data volume, velocity and data representation ability to
based on Relational database management system (RDBMS). perform effective analysis using traditional relational
There are two aspects in which RDBMS and Big Data approaches [15]. In March 2012, The Obama Administration
differs: announced that the US would invest 200 Million Dollars to
1) RDBMS can support structured data but big data supports launch a big data research plan [2].
for semi-structured and unstructured data.
2) RDBMS scale up to expensive hardware and cannot An IDC Reports predicts that from 2005 to 2020, the global
connect with commodity hardware in parallel and it’s not data volume will grow by a factor of 300, from 130
supported by big data. Exabyte’s to 40,000 Exabyte’s, representing a double growth
every two years[9]. IBM estimates that everyday 2.5
When does analytics become big data analytics? The size that quintillion bytes of data are created out of which 90% of the
defines big data has grown. In 1975 attendees of the first data in the world today has created in the last two years. It is
VLDB (Very large databases) conferences worried about observed that social networking sites like Facebook have 750
handling the Millions of data points found in US census Million users, LinkedIn has 110 million users and Twitter has
Information [8]. Big data analytics is the process of 250 million users [17]. From industry, government and
examining large datasets containing a variety of data type’s research community, Big Data has led to an emerging
Volume 4 Issue 4, April 2015
www.ijsr.net
Paper ID: SUB153031 845
Licensed Under Creative Commons Attribution CC BY
 International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
research field that has attracted tremendous interest. The 3. Big Data
broad interest is first exampled by coverage on both
industrial reports and public media for example: The Big data is the new term that contains large and complex
economist, New York Times [12]. Mobile Phones becoming datasets. It is difficult to manage these datasets without new
best way to get data on people from different aspect, the huge technology. The Mckinsey Global Institute (MGI) published
amount of data that mobile carrier can process to improve a report on big data that describes the various business
our daily life [13]. In figure 1, From Year 2005, it would opportunities that big data opens [12]. Paulo Boldi, One of
appear from this graph that the amount of data was the authors says “Big Data does not need big machines, it
practically increased. However, Consider exponential growth needs big intelligence” [6]. There are two types of Big Data
in data from 2005 year, when enterprise system and user is as follows:
level data was flooding into data warehouse [11].
3.1 Structured Data

These data can be easily analyzed. It is in numerical form,

figures, and transaction data etc.

3.2 Unstructured Data

These data contain complex information such as Email

attachments, Images comments on social networking sites.
These data cannot be easily analyzed.

Doug Lancy was the first one talking about 3v’s in big data
management [3]:
Figure 1: Exponential growth of data from year 2005 to Volume - It describes the amount of data. It refers to mass
2012[11] quantities of data.
Variety - It describes different types of data and sources
When the capacity of Data Warehouse grew from 50 GB to 1 including structured, semi-structured and unstructured data.
TB – 100TB. Data was in structured form when it creates Velocity - It defines the motion of data. Data created rapidly,
from many organizations. Data goes from three properties processed and analyzed.
like volume, Variety and velocity. Many companies were
facing the problem on how to expand the capacity of data
warehouse to accept the new requirement.

Figure 2 illustrates that there are variations shows in the

amount of data stored in different sectors by using the types
of data generated and stored i.e. whether the data is in audio,
video, images and text format and differ from industry to
industry[12]. Banking, Insurance and Health care sectors are
responsible for text/numeric data. Communication and Media
are highly responsible for audio and video type of data.

Sectors Video Image Audio Text/Numeric

Banking
Insurance
Retail Figure 3: 3v’s Big Data management
Wholesale
Utilities
Health care 4. Big Data Analytics
Transportation
Communication & Media Big data analytics enables organizations to analyze a mix of
Construction structured, semi structured and unstructured data in search of
Government valuable business information. Makinsey’s internal Think-
Education Tank, the Mckinsey Global Institute, published a major study
in June 2011 on Big Data [12]. Its overloading conclusion:
Penetration: Big Data is “a key basis of competition and growth”. The
term Analytics (including its Big Data form) is often used
Figure 2: Variations possible in generating and growth of broadly to cover any data-driven decision making [8]. The
data by using types such as audio, video etc.in various sectors term analytics divided into two groups: Corporate analytics
[12]. and Academic research Analytics. In Corporate Analytics,
Team uses their expertise in statistics and Data mining. In
Academic Analytics, Researchers analyze data to test
Hypotheses and form theories [8].
Volume 4 Issue 4, April 2015
www.ijsr.net
Paper ID: SUB153031 846
Licensed Under Creative Commons Attribution CC BY
 International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
In Big Data Analytics, Researchers found that the generated classification of mushroom as edible or poisonous[4]. It is
data divided into various Big Data application such as used for data mining.
follows [2].
5.3 Crowd Sourcing
4.1 Structured Analytics
A technique in which collecting data submitted by large
In structured analytics, large quantity of data is generated group of people or community i.e. crowd. It is usually
from business and scientific research fields. These data is through network media such as web.
managed by RDBMS, Data warehousing, OLAP and BPM.
Data grown by various research area like Privacy preserving 5.4 Data Mining
data mining, E-commerce.
A technique in which extracts patterns of data from large
4.2 Text Analytics datasets of combinations from statistics and machine
learning.
In Text analytics, Text is one of the most common forms of
storing the information and it includes Email communication, 6. Conclusion
documents, and Social media contents. Text analytics also
known as Text mining, refers to the process of extracting In this paper, we have presented the concept of big data. Big
useful information from large text. Text mining system is data is the large and complex datasets and it is generate from
based on text representation and Natural Language various sources like social media comments, playing a video
Processing (NLP) with emphasis on the latter [2]. game, email attachments etc. There is complexity in big data
such as velocity, variety and volume. These three terms are
4.3 Web Analytics more challenging for big data analytics. We have provided
literature survey shows exponential growth of data in
The aim of Web analytics is to retrieve, extract the industries from 2005 year. There are variations possible
information from Web Pages. Web Analytics also called while generating and storing data whether data is in audio,
Web mining. video, images and text. In big data analytics, Researchers
divided generated data into various big data application such
4.4 Multimedia Analytics as structured data analytics, text analytics, web analytics,
multimedia analytics and mobile analytics. Many challenges
Recently multimedia data, including images, audio, and video in the big data system need further research attention.
has grown at a tremendous rate. Multimedia analytics refers Research on typical big data application can generate profit
to extract interesting knowledge and semantics captured in for businesses, improve efficiency of government sectors.
multimedia data. Multimedia analytics covers many subjects
like Audio Summarization, Multimedia annotation, 7. Acknowledgement
Multimedia indexing and retrieval.
I would like to thank to all people who help me prepare this
4.5 Mobile Analytics paper completely. I would also thank to my guide who help
me and get proper suggestion. Finally I like to thank to all
Mobile data traffic increased 885PBs Per Month at the end of website and journal papers which I have refer to create my
2012. Vast volume of application and data leads to mobile review paper successfully.
analytics. Mobile analytics involves RFID, mobile phones,
Sensors etc.
References
5. Technique for Analyzing Big Data [1] Sameera Siddiqui, Deepa Gupta,” Big Data Process and
Analytics : A Survey”, International Journal Of
There are many techniques that can be used to analyze Emerging Research in Management & Technology,
datasets. Some techniques are machine learning. From this ISSN: 2278-9359, Volume 3, Issue 7, July 2014.
techniques, analyze new combination of datasets [12]. [2] Han Hu, Yongyang Nen, Tat Seng Chua, Xuelong Li,”
Towards Scalable System for Big Data Analytics: A
5.1 A/B Testing Technology Tutorial”, IEEE Access, Volume 2, Page No
653, June 2014.
A technique in which control group compared with various [3] Bharti Thakur, Manish Mann,” Data mining for big data:
test groups in order to determine what changes will improve A Review”, International journal of advanced Research
a given variable for example- Reponse rate of marketing. in Computer Science and Software Engineering, ISSN:
2277 128x, Volume 4, Issue 5, May 2014.
5.2 Classification [4] Anand V. Saurkar, Vaibhav Bhujade, Priti Bhagat and
Amit Khaparde,” A Review Paper on Various Data
A technique in which to identify the categories of new Mining Techniques”, International Journal of Advanced
datasets and assign into predefined classes for example- Research in Computer Science and software

Volume 4 Issue 4, April 2015

www.ijsr.net
Paper ID: SUB153031 847
Licensed Under Creative Commons Attribution CC BY
 International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438
Engineering, ISSN:2277 128X,Volume 4, Issue 4, April
2014.
[5] Puneet Singh Duggal, Sanchita Paul, ”Big Data
Analysis: Challenges and Solutions”, International
Conference On Cloud, Big Data and Trust 2013, Nov
2013.
[6] Albert Bifet, “Mining Big Data in Real Time”,
informatica, 2013.
[7] Stephen Kaisler, Frank Armour, J. Alberto Espinosa and
William Money,” Big Data: Issues and Challenges
Moving Forward”, Hawaii International Conference on
System Science, IEEE Computer Society, Page No. 995,
2013.
[8] D.Fisher, R.Deline, M.Czerwinski and S.
Drucker,”Interaction with big data analytics”, Volume
19, No.3, May 2012.
[9] J.Gantz, D. Reinset,” The Digital Universe in 2020: Big
Data, Bigger digital shadow, and biggest growth in the
far east”, in Proc : IDC iview, IDC Anal, Future, 2012.
[10] Denis Guyadeen , Rob Peglar,” Introduction to Analytics
and Big data- Hadoop”, SNIA Education Committee,
2012.
[11] Neil Raden,”Big Data Analytics Architecture”, Hired
Brains Inc, 2012
[12] James Manyika, Michael Chui, Brad Brown, Jacques
Bhuhin, Richard Dobbs, Charles Roxburgh, Angela
Hungh Byers, “Big Data: The next frontier for
innovation, competition and productivity”, June 2011.
[13] Wei Fan, Albert Bifet, “Mining Big Data: Current Status
and Forecast to the Future”, SIGKDD Explorations,
Volume 14, Issue 2.
[14] American Institute Of Physics(AIP), 2010. College Park,
MD(http:// www.aip.org /fyi/2010/)
[15] M.Cooper, P.Mell(2012). Tackling big Data(Online).
Http://csrc.nist.gov/groups/SMA/Forum/document/June2
012Presentation/f%CSM_june2012_cooper_Neul.pdf.
[16] www. Searchbusiness analytics.techtarget.com
[17] www.ebizmba .com/articles/social-networking-websites.

Volume 4 Issue 4, April 2015

www.ijsr.net
Paper ID: SUB153031 848
Licensed Under Creative Commons Attribution CC BY