Data Mining: Applications, Elements, And Challenges | Summary Statistics Of Accounting Jobs In Australia

Defining Data Mining and Its Applications

Part A

Data mining is simply the process of extracting unknown analytical data and information from big databases. Data mining can also be thought of as, procedures of evaluating data in large databases from diverse viewpoints and perceptions, classifying and condensing it into valuable information and knowledge (Data Mining Page 2017).  This revealed information is then used to increase sales and/or reduce costs for business organizations.

Data Mining Applications

Data mining is an emerging technology and is still being improved every day.  In spite of this, some business organizations continue to use this technology regularly (Ramageri 2010). Many   organizations including banking institutions, insurance companies, retail stores and hospitals use data mining. The following are applications of data mining in hospitals and business organizations

• In medicine, data mining has been applied to study the sequences of human genetics which helps in understanding DNA patterns. It is also used to study various cell patterns that may cause certain kinds of diseases.
• Data mining tools can be used to identify trends and patterns amongst customers and aid in identifying customers that are most likely to react to upcoming products and services for business organizations (Amazon Web Services 2017). This allows businesses to study their customers and therefore make smart advertising decisions in future.
• Data mining can be used to help human resource managers where they apply it   to identify features of the most successful employees. With better employees, a company is able to be more productive and hence increase sales and profits.

Elements of Data Mining

1. Extracting, Transforming, Loading transaction data onto the data warehouse systems

After data has been collected, it need to be structured in such a way that it becomes informative. It then requires to be converted it into useful information including size, category, time and quality before being stored in data centres. For example, job search information needs to be categorized into type of job and location of job for someone looking for a certain job in a specific locality after which it can be stored in warehouse servers.

1. Store and manage the data in a multidimensional database system

For data to be effectively analyzed, first of all, it needs to be collected and stored in data storage hardware or servers from where it can then be managed. For example in a student database, student information including names, age, course and grades need to be first collected and stored in database. From here it can then be analyzed for specific information including what course a given student takes, in which year are they and so on.

• Provide data access to business analysts and information technology professionals

Without proving collected, stored and structured data to data analysts and technology experts, it will deem useless. Therefore, after successful data collection and formatting, it is important to present to analysts so that they can deduce useful information from it. Only in this way can the data help businesses to make strategic decisions (Brown 2017).  For example, marketing analysts can be provided with customer data and information to be able to analyze customer tendencies to buy particular products and on which days. With such information, they can make decisions that would see a business increase its sales and improve customer feedback mechanisms.

1. Data analysis using data application software’s

To be able to deduce information from collected data, data analysts make use of data analysis software’s. In addition many large relational databases have data analytic and report generating tools embedded on them (Brown 2017). Some data analysis software include SPSS where data is entered and then analyzed for patterns and trends that can give important information.  In a hospital setting for example, researches can be able to use SPPS data analysis tool to be able to find out whether it is women, men or children that are largely affected by a particular disease.

1. Data presentation in form of  graphs or tables

The Five Major Elements of Data Mining

The last key element in data mining is the presentation of analysis findings. It is necessary to represent the analyzed data in clear understandable formats including graphs, tables, charts and so forth. When information is in form of rows and columns it may fail to make sense to a lot of people especially those who are not technical but after it has been presented in graphs and charts formats then a lot of people are able to make meaning out of it. For example, analysis for mobile phone sales can be presented in a table format where, the highest sales of a given phone model is o top and the lowest model type sales at the bottom. The same information can be resented in a pie chart where the large percentage sales would represent the highest sold model type and vice versa.

Data mining has become a very popular technology for analyzing data in business organizations. Business are now able to classify , categorize and analyze it for patterns and trends and are able to get information that increase profits and sale targets which overall enhances business productivity. However, data mining itself comes with its own share of challenges according to an article by Big Data Made Simple (2015). First, Data mining incurs high costs of installation and maintenance. Installation of high quality database servers such as highly relational databases and quality storage servers and hardware requires a lot of money. The databases also require expert administrators which also cost the organizations. Second, the proliferation of data and information has become very complex. This is from the dawn of synchronization of technologies including email, social media that is easily available for millions of people. It is not easy to manage such quick in streams of data. Third, the data to be analyzed requires to be structured because on collection, it is usually unstructured. In addition, the data could be collected in form of diverse formats which could be incompatible with each other. Fourthly, data set to be analyzed is not static and is usually unbalanced. With every minute, there is new data inflowing in which makes the process of categorizing and analyzing the already captured data complicated. Lastly, developing quality data analysis algorithms that will efficiently and effectively perform data analysis is very difficult. Some algorithms may fail to analyze data successfully meaning, that the result may not be worthwhile.

1. 20, 371  available accounting  jobs
2. The table below shows job classifications with the highest number of jobs in an ascending order

Challenges in Data Mining Analysis

S. No	Category	Number of Jobs
1	Sales	3202
2	Administration & Office Support	1575
3	Information & Communication Technology(ICT)	1008
4	Banking & Financial Services	798
5	Real Estate & Property	602
6	Health Care & Medical	547
7	Call Center & Customer Services	497
8	Marketing & Communications	484
9	Manufacturing, Transport & Logistics	436
10	Human Resources & Recruitment	398

Table 1: Source: Seek Limited (2017)

From the table above, it is clear that sales jobs are the highest with a figure of 3202 followed by Administration & Office Support jobs with a figure of 1575.The third highest job category is Information & Communication Technology with 1008 jobs, fourth highest is Banking & Financial Services with about 708 jobs , followed by Real Estate & Property with 602. The sixth highest is Health Care & Medical category with around 547 jobs, followed by Call Center & Customer Services with about 497. Marketing & Communications category follows with 484 jobs, then 436 jobs in Manufacturing, Transport & Logistics and lastly Human Resources & Recruitment with around 398 jobs.

1. The table below shows the amount paid between four  particular category payments and the number of jobs found in the category

S. No	Amount	Number of Jobs
1	$30k – $40k	177
2	$50k – $60k	2001
3	$70k – $80k	2103
4	$80k – $100k	2330

Table 2: Source : Seek Limited (2017)

From the table 2 above, we can deduce that jobs paying between  ($80k to $100k) annually are the highest with a figure of 2330 while the low paying category ($30k to $40k) annually are low with a figure of 177 jobs. Therefore, the higher the salary amount per year, the higher the number of available jobs and vice versa. It is also important to note that from the table that there is no much difference between the number of jobs in the salary scales between $50 k and $60k and that one between $70k and $80k. The difference in the number of jobs between the two scales is only 102 which is not a huge difference compared to the others with a difference of 874 jobs between the $30k and $40k and $50k and $60k.

The table below number of Accounting  jobs in the major Australian states

S. No	State	Number of Jobs
1	Adelaide	217
2	Melbourne	1588
3	Sydney	2536
4	Perth	450
5	Hobart	20
6	Brisbane	715

Table 3:Source : Seek Limited (2017)

 From the table, accounting jobs are highest in Sydney Australia with 2536 available jobs and lowest in Hobart with only about 20 Accounting jobs. Sydney state therefore is most likely the place where accounting graduates would get jobs.  Another with a high likelihood of accounting jobs is Melbourne with about 1588 jobs available. Other states where accounting graduates can easily find accounting jobs include Brisbane and Perth with about 715 and 450 jobs respectively.

1. Summary findings

The data mining results above present job category with the highest available number of jobs. Sales jobs are the highest in number while Human resource and recruitment has the lowest number of jobs. Specifying the accounting category, jobs with the high annual salaries are higher than those with lower annual salary of about $30k to $40k. This shows it is possible for an accounting graduate to find a high paying job with a salary of about $80k to $100k per year. Accounting graduates are likely to get jobs in Sydney since that is the state with the highest number of accounting jobs followed by Melbourne with about 1588 accounting jobs. Hobart state has the lowest number of accounting jobs.

References

Amazon Web Services, 2017, 5 real life applications of Data Mining and Business Intelligence, [Online], Available: https://www.matillion.com/insights/5-real-life-applications-of-data-mining-and-business-intelligence/ [Accessed April 20 2017].

Brown, M, 2012, Data mining techniques, [Online], Available: https://www.ibm.com/developerworks/library/ba-data-mining-techniques/ [Accessed April 20 2017]

Big Data Made Simple, 2015, 12 common problems in Data Mining, [Online], Available: https://bigdata-madesimple.com/12-common-problems-in-data-mining/ [Accessed April 21 2017].

Data Mining Page, 2017, An Introduction to Data Mining, [Online], Available: https://www.thearling.com/text/dmwhite/dmwhite.htm [Accessed April 21 2017].

Ramageri, B, 2010, Data mining techniques and applications, Indian Journal of Computer Science and Engineering, Vol. 1 No. 4 301-305 [Online], Available: https://www.ijcse.com/docs/IJCSE10-01-04-51.pdf [Accessed April 20 2017].

Seek Limited, Job Search, 2017, [online], Available: https://www.seek.com.au/ [Accessed April 21 2017].