Good Data Mining Essay Example

Data mining is a computational process that involves analysis of data in order to uncover hidden patterns that tend to deviate from the norm. This deviation is used to generate a predictive analysis of the data (Abbott, 2014). The data is retrieved from sources such as databases, and a data mining application software is used to analyse data. Data mining is the new modern way of analysing data especially in the business sector.
My case study involves Company A, that deals with the supply of electrical gadgets to various huge corporations. The company supplies tablets, phones and desktop computers to business corporations. They also specialize in configuring such gadgets to be in conformity with the data network of those corporations. Company A has been in operation for six years from 2008-2014 and has established a customer base of 134 loyal corporations that mainly award them with tenders to supply electronic gadgets. The information Technology (IT) department of the company A has kept records of all the dealings and financial transactions between the company and its customers in an readily available database. Recently Company A has not been doing well, and the IT department has strategized to use data mining analysis to get to the root of the problem as they deal with many customers.
The primary reason for the selection of data mining technique is that it is not time-consuming and mostly automated. It would, therefore, provide the company with a fast solution incorporated with a predictive analysis of the data retrieved from their database management systems. Furthermore, it would save on costs as it does not involve hiring more workforce experts in the IT department of Company A. The Company intends to use the famous Dell’s company data mining software application known as “Statisca”. The software is very user-friendly and is a commercial-off-the-shelf product. Any computer literate person who knows how to use Windows applications such as Microsoft Excel or Word can easily use the application without any difficulties (Miner, 2012). It can also be customized to fit in the company’s culture, and industry standard scripting languages such as Visual Basic or R are used for configuration of the software application. “Statisca” objects can be called by COM compliant programming languages such as C++, C# and Java.
Company A would have opted to use the traditional Exploratory Data Analysis (EDA). However, this system in the present time is currently not very efficient in providing data mining solutions. Exploratory Data Analysis is divided mainly into Hypothesis testing and Graphical (data visualization) techniques. Hypothesis testing is designed to authenticate a priori assumptions. This technique does not bring out any predictive analysis that is concrete.
The Graphical (data visualization) EDA technique is effective to some extent. It is an interactive procedural method that allows one to pick on-screen specific data points or subsets of data and identify their similar characteristics. It also scrutinizes their effects on relations between relevant variables. This is essential in data mining but does not serve the higher purpose by coming up with a predictive analysis that give Company A, a clear guide towards the direction to follow. Although data mining borrows heavily from the traditional principles of EDA, the major difference is that it is more oriented towards applications than the fundamental nature of the underlying phenomena (Foreman, 2013). It is less concerned with identifying precise relations between the variables in question. Data mining is, therefore, the best choice of data solution for Company A as it is modern and very accurate in determining unusual patterns in the data collected.
The statement of the question involves how Company A would revive its business that seems to be not doing well presently as compared to the past. There has been a major decline in demand supply of their products, and they are losing some of their major clients. The customer base of Company A has reduced to a whopping 82 after six years of business. This is in contrast to the 134 customer base that was obtained in 2010, two years down the line after starting up the business. The data mining analysis is meant to show a company A the significant unusual trends in the sales database system of the enterprise in order to determine the cause of the continued failure of the company presently.
The data mining analysis solution was implemented by the IT department of Company A and within two weeks a course of action decision was made based on the predictive analysis provided by the competent data mining software application “Statisca”. It was noted that in late November 2009 when the company was at its peak performance, the Human Resource Department was recruiting many persons to work as delivery guys in the business. After the huge recruitment of these persons, the company slowly started losing business clientele. Most of these employees were not thoroughly trained in dealing with customers in the corporate world ; hence they ended up, not being competent in their work. There were several complaints by customers of delays in delivery and supply of faulty products which seem not to have been taken seriously by the Customer Services Department of Company A. The predictive analysis through data mining led to the conclusion that Company A had began failing due to poor handling of customer’s issues by the employees hence most clientele felt that they did not get quality services that was equivalent to the value of their money. Company A was also to change its manufacturing supplier and also raise a complaint on the stock of faulty products that have cost them their business.
Another data source method that would adequately answer this statement question is by conducting a market survey on their customers. This would actually collect the data on issues affecting their clientele. It would give Company A an opportunity to be well aware of the standards required by their customers and be able to correct on all the issues arising from their customers in due time. Customer satisfaction would be a priority and Company A would jumpstart back to a thriving business with a continually growing customer base.

References

Abbott, D. (2014). Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst. Hoboken, New Jersey: John Wiley & Sons.
Foreman, J. W. (2013). Data Smart: Using Data Science to Transform Information into Insight. Hoboken, New Jersey: John Wiley& Sons.
Miner, G. (2012). Practical Text Mining and Statistical Analysis for Non-structured Text Data Applications. Waltham, Massachusetts: Academic Press.