Exploring Security Challenges in Next-Generation Big Data Analytics

This literature review investigates the security challenges in next-generation big data analytics, focusing on infrastructure support applications. It examines various data types, including online network data, mobile and IoT data, geography data, spatial temporal data, and streaming real-time data, highlighting the unique challenges each domain presents. The review discusses threats such as adware, spam, unlicensed spectrum usage, power consumption attacks, and manipulation of geographic data. It also addresses challenges irrespective of the domain such as noise accumulation, spurious correlation, high dimensionality, computational cost, and heterogeneity. The research methodology employed involves both qualitative and quantitative data collection from peer-reviewed journals, as well as primary data collection through surveys and interviews with IT professionals in data analytics. The common challenge identified is the need for robust infrastructure to support machine learning algorithms and ensure efficient security in next-generation big data systems.

ADVANCED RESEARCH TOPIC: Next-Generation Big Data Analytics:
State of the Art
RESEARCH QUESTION: What is the common challenge in infrastructure-support
applications in terms of efficiency in security for next-generation big data analytics?
ABSTRACT:
The term big data is the buzzing word in present days. The greater number of subjects we use
everyday life to basic research fields, which involves big data problems. In this research
study we review the most popular data types, data security, privacy and storage models
related to network big data. The data types we use for our study are online network data,
mobile and IoT data, geography data, spatial temporal data and streaming and real-time
data. Various kinds of domain give particular challenges which are properly estimated, and
important impact is created by next-generation of big data systems. The increased
performance in the building designs of big data are improves by the effect of functionality
and special interest in big data systems of next generation. The safety and protection of big
data is a very vital problem. The analysis big data can be done by the MapReduce-related
method. Also, in this study we have explained the applications of big data as transient power
prediction, content recommendation, smart city and user behaviour prediction. The common
challenge is that the infrastructure is vital to support machine learning algorithms to
implement new ones.
Key words: Big data, Machine learning algorithm, Data types, and challenge.
Literature Review
According to Fan, Han and Liu (2014), Online Network Data has been one of the
major data set that is to be used for understanding the challenges that are to be considered is
Online Networking data. This section ensures that data that are generated via the platform of
Facebook and Second Life are considered in this data set. Adware and Spam are the major
threats that are present in the online network data. Due to occurrence of these attacks data
might not be protected and hence wise collection of data and hence wise analysis of the same
gets affected in a negative manner. Another major trend that is seen is data that are collected
from the mobile and IoT data [5]. These data are also considered as a major data set that is
used by the Next generation Bug Data in the analysis process [9]. There is a major threat that
has been affecting the data analysis section. Integrated mobile webs have been using
unlicensed spectrum. This leads to the fact that not all the fata that are used by the users can
be gathered. Usage of cognitive radio technology has been affecting the entire data collection
and analysis process. Power consumption attacks has been another threat that has been
affecting the analysis process. Structuring of the data has also been a major problem in the
functional process. Geography data is also considered as a major data set that helps in
performing proper data analysis via next generation data analysis [2]. Despite the fact that
this data sets are in process of getting collaborated with the OSN, the data itself has
importance in the data analytics section.
It has been seen that researchers have been using unprocessed data sets. This kind of
unprocessed data set includes the fact that the subsets can be created and hence wise
replication of the data can be performed. Another major issue that can be stated is that

collection of this type of data will be requiring proper knowledge in geography as well as in
topological section. Hence a proper blend of topological as well IT informatics acts as a
major challenge in the analysis process. This might lead to inaccurate data analytics.
Manipulation of geographic data has been another issue as well. Spatial temporal data type is
another data type that acts important in the data analytics section. In this case data collection
acts as a main challenge [7]. The collaterals that are present in the path of spatial temporal
data acts as a major threat to the data collection process. Data are collected as per the sensors
and the security process might be fragile in this kind of data collection process. This leads to
the fact that data collection and data analysis might be getting affected. Another data type that
is considered is Streaming and Real time data. In this case the major issuer that is present is
that implementation of firewall cannot be performed [1]. This leads to the fact that with the
presence of firewall, detecting of anomalies was easier and without the presence of the
firewall the entire process gets affected in a negative manner. These are the main threats that
might be affecting the functional process of the data management. However real time data
analytics itself acts helpful in detection of the threat. As with the help of the real time data
analytics, a holistic view of the network can be established. This leads to the fact that in case
of any anomaly in the network, detection of the anomaly gets easier. Real time visual data is
also acts as an important data type that has been providing enough data to the next generation
big data for the data analytics section [6]. There are a lot of challenges that are faced in the
collection and analysis of the data that are generated via the visual data. Analysis and
retrieval of data has been a major challenge in the process of data collection and assessment.
The main aspect that is to be considered is the retrieval of the data. Analysis of the data has
been one of the major issue that is to be faced. Presence of weak semi supervised deep
learning is also another issue that is to be considered. Image annotation process is negatively
affected due to this terminology [3]. Weak labelling of images might be acting as a major
issue. Presence of triplet similarity loss has also been acting as a major challenge and threat
in case of the visual data collection process. It can be stated that the data that are weakly
weighted pairwise ranking loss is yet another issue that is seen in the faced by the next
generation big data.
There are a certain challenges irrespective of the domain that effects the analytics of
data with the help of next generation Big Data [10]. The major challenges includes the likes
of presence of noise accumulation and hence wise spurious correlation as well as the
incidental homogeneity also occurs due to the presence of noise accumulation. Another
challenge that is faced includes the likes of having high dimension along with the combined
large sample size. This leads to creation of challenges such as instability in the algorithm of
the entire program. Another issue that was faced includes the likes of high computational
cost. Heterogeneity is also another challenge that is present in performing data analytics via
the platform of new generation big data [4]. These heterogeneity causes the statistical biases.
This statistical biases leads to proper assessment of the adaptive and robust processing. It is
seen that the adaptive nature of the entire process has been very low. Despite the issues
present there are many recommendations that can eliminate the issues and enhance the data
analytics process. Performing of statistical accuracy has been performing proper assessment
of statistical analysis and hence wise higher accuracy can be established [8]. This process
tackles the functioning of the noise that are accumulated in data and hence wise elimination
of noise accumulation can be performed. This will increase the accuracy in the data analytics
section.

Research methodology
The research methodology plays an important role in the processing of the research
project. Research methodology is mainly focussed on detailing the different steps that are to
be incurred in the execution of the project. The strategies that are implemented for collection
and analysis of the data are considered in this section. The initial step of research
methodology is to perform a research strategy. This research strategy is mainly responsible
for performing the assessment of the research methodology. As per this section the stages that
will be performed in the research methodology are performing the qualitative and quantitative
data collection [13]. After performing this section, the approach that is to be taken by the
researcher is decided after performing the qualitative and quantitative data collection and data
analysis technique. As it has been decided that usage of inductive approach can be performed.
As per the research strategy, data collection will be made via tools which will be increasing
the efficiency of the data collection process. After this process according to the research
strategy formulation, selection of sample will be performed. In this case purposive random
sampling method will be used [12]. After this section primary data collection will be
performed. After this section the ethical dilemmas will be entertained and hence wise
elimination of the ethical constraints will be made.
Developing of a research approach is made. With the help of the approach the main
aspect that is considered is that the entire research process narrows. This section helps in
proper understanding of the exact goal where the research will be focussed on. After this
section the secondary data is collected. In this case as per the research strategy that is stated
above, secondary data collection will be made. In this case qualitative data collection is
performed. The very first step that is taken is to perform qualitative data collection. Literature
review is the major consideration of the process [11]. Literature review helps in proper
discussing about the data has been already generated by the scholars. Implementation of the
data generated by the scholars ensures that the accuracy of the data that is being used in the
project is high. Another reason of using the data collection method is that the data that is
collected is highly authorized. The data that are used are properly referenced. Many research
papers suffers from the copyright issue. In case proper peer reviewed journals are made,
copyright issue will not arise. This will ensure that the data that is collected will be having
high authentication as well. With the help of the secondary qualitative data collection the
data, conceptualization of the entire topic can be performed. With theoretical data collection
proper understanding of the basics of the research can be performed. After performing the
secondary qualitative data collection, quantitative data collection is made as well. This
quantitative data collection is also done form the peer reviewed journals [14]. Journals acts as
the major source that is used. An aspect that is considered is that numeric data collection
helps in increasing the preciseness and decisiveness of the report. This phase of the research
acts as one of the most important stage that is to be performed. After performing this stage a
basic understanding of the topic is made. Hence wise this process ensures that the data
segmentation is performed. After performing of the secondary data collection. After this stage
primary data collection will be performed. In this process both qualitative as well quantitative
data is collected. There are 2 major type of data collection process. The process includes the
likes of conducting surveys and taking interviews. This conducting of surveys along with the
interview will be performed in a complete different manner. The data that are collected in the
interview is qualitative in nature. Again the data that is collected as per the survey resultant

will be quantitative in nature. For conducting the survey sampling is to be performed in order
to select the sample. In this case random sampling will be made. The reason of using random
sampling is that biasness in the result that is received is very low. In this case employees of
IT firms are considered who are mostly into data analytics will be accepted. After this process
random sampling will be made. A set of question will be prepared. The questions will be
mostly emphasised on gathering numeric and concise data. This section will be used for
increasing the preciseness of the research project. The question set that will be provided will
be provided to the survey takers. The resultant that will be received helps in understanding
the trend. After conducting the survey another aspect that will be considered is the interview
that will be given [15]. The interviewees are often managers of different IT firms. The
interviewees will be experienced in the field of Big Data and analytics. Primary qualitative
data is collected in the process. The interview questions are generally descriptive to answer in
nature. Hence a separate set of question will be required for performing the data collection
and analysis section. The main reason of this session will be descriptive and qualitative in
nature is that the data that will be collected from this interviews are highly accurate and hence
wise these data will be highly accurate in nature. After collecting the data proper analysis of
the same is also performed. A blend of qualitative and quantitative data is performed the
process. The data that will collected are documented and hence the research process will be
get continued. After this data collection and analysis, proper assessment of the ethical
considerations is also made. In this section the main aspect that is considered is that in
performing the entire process is there any kind of ethical issue. In case there is no ethical
constraint, the entire process will get started [12]. This will ensure that the data that are
collected are authenticated and hence wise they can perform the research process. On
following the above stated steps the project will be getting completed in an efficient manner.

References
[1] J. Fan, F. Han and H. Liu. Challenges of big data analysis. National science review, 1(2),
pp.293-314, 2014.
[2]A. Labrinidis and H.V. Jagadish. Challenges and opportunities with big data. Proceedings
of the VLDB Endowment, 5(12), pp.2032-2033, 2012.
[3] A. Katal, M. Wazid and R.H. Goudar. August. Big data: issues, challenges, tools and
good practices. In 2013 Sixth international conference on contemporary computing (IC3) (pp.
404-409). IEEE, 2013.
[4] L. Manovich, Trending: The promises and the challenges of big social data. Debates in
the digital humanities, 2, pp.460-475, 2013.
[5] M.M. Najafabadi, F. Villanustre, T.M. Khoshgoftaar, N. Seliya, R. Wald and E.
Muharemagic. Deep learning applications and challenges in big data analytics. Journal of Big
Data, 2(1), p.1, 2015.
[6] R. Akerkar. Big data computing. Crc Press, 2013.
[7] E. Bertino. Big Data--Opportunities and Challenges Panel Position Paper. In 2013 IEEE
37th Annual Computer Software and Applications Conference (pp. 479-480). IEEE, 2013.
[8] L. Wang, G. Wang and C.A. Alexander. Big data and visualization: methods, challenges
and technology progress. Digital Technologies, 1(1), pp.33-38, 2015.
[9] H.V. Jagadish, J. Gehrke, A. Labrinidis, Y. Papakonstantinou, J.M. Patel, R.
Ramakrishnan, and C. Shahabi. Big data and its technical challenges. Communications of the
ACM, 57(7), pp.86-94, 2014.
[10] U. Sivarajah, M.M. Kamal, Z. Irani and V. Weerakkody. Critical analysis of Big Data
challenges and analytical methods. Journal of Business Research, 70, pp.263-286, 2017.
[11] C. MacDonald. Understanding participatory action research: A qualitative research
methodology option. The Canadian Journal of Action Research, 13(2), pp.34-50, 2012.
[12] D. Tuohy, A.Cooney, M. Dowling, K. Murphy and J. Sixsmith. An overview of
interpretive phenomenology as a research methodology. Nurse researcher, 20(6), 2013.
[13] T.Roberts. Understanding the research methodology of interpretative phenomenological
analysis. British Journal of Midwifery, 21(3), pp.215-218, 2013.
[14] A.M. Novikov and D.A. Novikov. Research methodology: From philosophy of science
to research design. CRC Press, 2013.
[15] T. Miller, M. Birch, M. Mauthner and J. Jessop. Ethics in qualitative research. Sage,
2012.