Report on the Big Data Analytical

CONTENTS:-

CONTENTS:- ............................................................................................................................................................ 1
INTRODUCTION ................................................................................................................................................... 1
RESEARCH ISSUES ................................................................................................................................................... 2
Literature Review:- ................................................................................................................................................. 3
Methodology:- ........................................................................................................................................................ 6
PROCESS: research and system development ..................................................................................................... 6
METHODS: DATA COLLECTION ........................................................................................................................... 6
Data analysis ...................................................................................................................................................... 7
Conclusion ............................................................................................................................................................... 7
References: ............................................................................................................................................................. 8

INTRODUCTION

The new revolution in data analytical has arrived, now its the age of Big Data. As the

traditional Data Analytical is not able to handle data in large quantities, the new technology

needs to be introduced for to dig in the Big Data and bring out the useful information. There

are the processes which includes high profile platform to analyse of the large data and the

data mining algorithm to cover the hidden processes which are stored in the Big Data. The

Big data give the benefits to the different industries like in Media Organisations and financial

organisations. Along with the number of benefits, big data technology also has the demerits

(Big Data Literature Search, 2014).

RESEARCH ISSUES

Defining Big Data: The main issue is defining the word Big Data. Big Data is referred to the

integration of the Organisational data at one place either in the super computer or in the cloud

system. Due to integration of all the data in the organisation it will be easy to get the right

information from the whole data for the decision making process.

Speed Issue: The different procedures are being issued in the mining of the data from the Big

Data. As it takes much time to explore the data from the huge storage of data.

Manipulation in the Business: Another issue is the use of the tools in data mining from the

big data, as most of organisations dont have proper skills to mine the data. The organisations

dont use the proper tool for abstracting the data from the Big Data. Many companies have

deploy the data warehouse techniques in regard to the storage of the data and now with the

new technology arrived, the companies have to clear the existing infrastructure to make the

room for the new technology to be deployed.

Clouding Issue: Another big issue in Big Data is storing the data in the cloud system, as in

the Big Data the data can be in the range of Terabytes and it may take more time than the

usual to store the data in the cloud system.

Knowledge to the Technology: To run the Big Data technology, the employers of the

particular organsiation must be skilful with three major skills that is business skill,

Information technology skill and Statistical skills.

difficult to retrieve data from the Big Data. If some of analysis is offline, the data inside the

Big Data must be played and could be retrieved by some other way.

Privacy Issue: In the Big Data, data is stored in the huge amount either in the super computer

of the company or in the cloud system. Sometimes the personal information is also attached

to the file to know the file description, which can be access by any user who is using the Big

Data to receive the data which may make a Privacy problem as the personal information can

be stolen.

Outcome as required meaning: The result of the Big Data must be according to the

requirement. As there is numbers of files of data are stored in the database, so it is necessary

that the information that is required must be driven from the same data file.

Literature Review:-

The new revolution in data analytical has arrived, now its the age of Big Data. As the

traditional Data Analytical is not able to handle data in large quantities, it arises the new

question that how the data can be analyze effectively from such a big storage data and how

the technique will be design to mining the data from the data. in this paper the author gives

the introduction to the Traditional Data Analytics and the Big Data Analytics. The issues

regarding the Big Data is also discussed in this paper and some open issues and research

which is done on the Big Data is also discussed (Wei Tsai and Feng Lai, 2015).

In this paper, the author describe about the developing of the system Risk Model which will

concern the systematic interrelationship of the financial institution with the risk management.

The author describes that the model will not only store the information about the financial
Report on the Big Data Analytical
market prices but also the data coming from the Financial advisors or states will also be

record in it to make it Big Data Analytical. The main purpose of writing this paper is to bring

the two different data sources, Financial Market and Financial Tweet together by the process

of Bayesian approach. It is described in the paper that how the big data s being introduced to

present Systematic Risk Model and how it will work with the combination of the two

financial sectors (Cerchiello and Giudici, 2014).

Big Data Analytics as well as Deep Learning are two main concerns in the field of science

technology. The private and public sectors have started using the Big Data to store the

necessary information which may helpful in storing the information about the marketing and

security threats and information of other sectors. The big IT Companies are also analysing the

use of the Big Data in the future which have positive impact on the business analysis and the

decision making process about the future technology. Deep Learning is an algorithm used to

extract the information from the complex data through Hierarchical Learning Process. The

complexity of the data is being evaluated at the starting process of the algorithm. Deep

Learning is a useful tool to extract the information from the complex data where the data is

not labeled and un-categorized. In this paper, it is described that how the Deep Learning can

be a good tool in the Big Data Analytical in terms of addressing the problem of the Big Data

regarding the data retrieve process. Some aspects are also investigated on Big Data about the

exploration of the Deep Learning research that would include streaming data , distributed

computing etc (Najafabadi and Villanustre, 2015).

The main purpose of this paper is to introduce the different platforms which can be used for

the Big Data Analytical. In this paper the merits and demerits of the different hardware

platforms have been discussed which will indicate which hardware platform is best for the

Big Data Analytical. In this paper the platforms are evaluated on the basis of scalability, time
Report on the Big Data Analytical
taken in process, the rate of data input and output. The software frameworks are also

discussed in the paper along with the hardware platform for the Big Data for their merits and

demerits. The two hardware platforms and the software infrastructure are compared with the

Qualitative method to get the righteous information about the two platforms and software in

which the six characteristics has been chosen from the data analysis and the platforms and

software infrastructure were rating using the Star rating Table. In this paper the K-means

clustering algorithm has been used on the each of the platform to get more information about

the working of the platforms in matter of the Big Data Analytical (Singh and Reddy, 2014).

Data streams is the word used in the term of buffering of the audio or video over the

channel in continuous flow. The streaming of the files is often unstructured which make it

difficult to store and process with the earlier simple technique. The Data streaming mainly

has four challenges which make it difficult for the traditional Data Analytical to process the

data. Those difficulties are the length of the stream which is countless, the drfit in the concept

which happen due to slow changes and evolution in the feature with the time of merdia

growth. The main reason for the occurrence of the Concept-evolution is the unknown class in

the data. There is another concept of Feature-evolution which occur due to the changes in the

old features to make it better and excited. The knowledge to the concept is very essential in

term to perform any big Data Analytical. In this paper, the research from the various

researchers has been evaluated. In their research the main objective was to conduct research

for the problem of big length data streaming and concept drifting. In this paper various string

based methodology has been described which are very effective to process data streaming

and the cover the challenges like infinite length and concept drift (Chandak, 2016).

The manufacturing industry is the new field seeking for the help of Big Data in the

organsiation to set the record of the manufacturing goods and the demands in one chart so
Report on the Big Data Analytical
that the decision making process would become easy. The manufacturing industry wants to

driven from traditional data analytical scheme to some more effective techniques. In this

paper, it is described that how the new techniques will be concentrated on merging the real

time data with the manufacturing intelligence for the timely accurate decision making

process. In this paper it is realize that the various technologies like IOT and CPS will be

emerged to measure the real time data process in the industry. With the use of these

technologies the demand as well as the production level in the industry will increase at high

speed. The production speed of the product will be at new height which may need numerous

of data to be processed and controlled the manufacturing unit. Hence the manufacturing units

will be enable to manage the increment in the demand and will use the analytical techniques

to get exact information for the production unit. This indicates that the big manufacturing

industries may need Big Data Analytical technology for this smart production. In this study,

the author stated that they used the simple research method that is systematic mapping or

Hierarchy process to show the process of the Big Data technology in the manufacturing

industry (ODonovan and Leahy, 2015).

Methodology:-

PROCESS: research and system development

There are two methods which can be used to do the research upon the Big Data technology.

In Observational Method various aspects of the Big Data is observed to derive the required

result from the research. In Qualitative Method, those methods are used to do the research

which are Quantitative by nature. These methods are quiet effective in getting the required

information to know the issues and problems regarding the Big Data technology.

METHODS: DATA COLLECTION

1. Questionnaire Methods

2. Interviews Methods

3. Observations

The data collected by these methods is appropriate because this data is collected by the real

time experience of the companies and individuals who are working on this technology which

is accurate and contains all the aspects of the issues of Big Data technology.

Data analysis

To analyse the data for getting the exact information, Observation method is used for its best

approach. It is useful because it provide the exact information which is necessary for the

decisions making process.

Conclusion

In this file we learn about the new technology that is Big Data technology which is used to

store the bulk data in the single place and all the data is integrated with each other so that the

necessary information can be derived from the combination of the data. The various

techniques have been discussed which can be used to derive the data from the Big Data

technology such as Data Mining Algorithm. The Big Data Technology has the big benefit to

the organisations which are carrying the bulk of data to be stored and retrieve the useful

information such as risk management and Multimedia organisation which has the numerous

of file to be stored. There are some issues with the Big data regarding the speed of data

searching and techniques used to search the data. , the research from the various researchers
Report on the Big Data Analytical
has been evaluated. In their research the main objective was to conduct research for the

problem of big length data streaming and concept drifting. In this paper various string based

methodology has been described which are very effective to process data streaming and the

cover the challenges like infinite length and concept drift.

References:

Wei Tsai, C. and Feng Lai, C. (2015). Big data analytics: a survey. Journal of Big Data.

Cerchiello, P. and Giudici, P. (2014). Big data analysis for financial risk management. Using Big data for

forecasting and statistics.

Najafabadi, M. and Villanustre, F. (2015). Deep learning applications and challenges in big data analytics.

Journal of Big Data, 02(01).

Singh, D. and Reddy, C. (2014). A survey on platforms for big data analytics. Journal of Big Data, 02(08).

Chandak, M. (2016). Role of big-data in classification and novel class detection in data streams. Journal of Big

Data, 03(05).

ODonovan, P. and Leahy, K. (2015). Big data in manufacturing: a systematic mapping study. Journal of Big

Data, 2(20).

Big Data Literature Search. (2014). Big Data, 2(4), pp.230-232.