Data Warehousing and Data Mining

Data Warehousing and Data Mining

Definition and Use

Data mining can be defined as the extraction of information that has been automated and has been predicted (Thearling, 2009). In other words, it involves analyzing data from diverse angles and dimensions, putting then into categories then summarizing the recognized relationships.

Data mining is now used in the security department. Firms, stores and other businesses are using this to the advantage of preventing theft as opposes to CCTV (Closed Circuit Television) and EBS (Electronic Article Surveillance), which are used to catch the thieves in the act (Hadfield, 2009). They are also using the application to cut costs and increase revenue.

Jager, a clothing chain store business based in Britain, started using this application in the summer of the year 2008. This was in an effort to es6tablish the company’s cause of loosing so much money. The business used the application to get employees who are involved in fraudulent or thieving practices. Working practices that lead to stock wastage can also be identified. The business expected to make an ROI (return on investment), within the first 6 to 9 months after the system went live. However, a number of difficulties were been encountered as the business had already existing complex applications. Making them, work in coordination was quite a challenge (Hadfield, 2009).

Analytical Tools Used

            There is software that is purchased, which is centralized in order to receive any data fed from any source as long as it is connected to it. In Jeer’s case, they got the software from IDM software, although any other type of software can be accessed. Before a full report is given, businesses ask the application more questions in order to separate a genuine loss and a false positive. Each question is phrased in relation to the answer of the question asked before. This application is empowered if it uses a larger number of data fed by the other systems. For example, the CCTV could be used to interpret data from Epos (Electronic point of sale). Businesses with RFID (radio frequency identification) projects could make use of the data from tagged pallets or items that are tagged individually within the data mining applications.

Auditors use a kind of network theory known as link analysis in order to understand the data patterns in the diverse systems. They search for patterns that are symmetric between one set of data and another. They also compare the data with asymmetric patterns in order to get an understanding of the relationship in the information.

Role of data mining and warehousing

            Warehousing has a number of roles. It performs disk or server bound functions related to asking and reporting on disk or servers not needed by processing systems in transactions. It is also involved in the use of server technology and data models that make the questioning and reporting process faster. It also tackles those that are not suitable for processing transactions. Warehousing provides an atmosphere where a suitable quantity of knowledge of the technical parts of database technology is needed. This is in order to write and preserve the questions and reports. It gives a means to fasten the writing and preserving of the questions and reports by personnel in the technical departments (Greenfield, 2005).

Warehousing also provides a “storehouse” of good procession of data in transactions that can be reported against them. The process must not necessarily need fixing the systems. Another role-played is that it makes questioning and reporting data from many systems much easier on a regular basis. It is also an advantage to the data from outside sources and that, which must be stored for question or report purposes alone. Warehousing has a role of acting as a “storehouse” of data in the transaction processing system. This contains data from a bigger period than can efficiently be held in a system. It also stops people who only need to question and report transaction systems data from gaining any access to these system databases and the judgment used to preserve the databases (Greenfield, 2005).

Data mining is used in research marketing and understanding the consumer behavior. As mentioned earlier, it is used to detect fraudulent activities from the employees, which greatly affects the business’ income and revenue. It is has a major role to play in ecommerce, health industry, the customer relationship management, just to mention but a few. Its ability to compile and analyze the different types of data coming in is greatly used in the business world (Thearling, 2009).

Warehousing Concepts

            The dimensional data model, the conceptual, logical, slowly changing and the physical data models are the most common concepts in warehousing. This is an advance state of the third normal form model. The data is stored in different forms in both models. In the slowly changing dimension, the recording attribute differs over time. This makes it sensitive to work with. The relationships of the highest level between the entities are identified in the conceptual data model. In this concept, there is no specification of attributes and primary keys. The logical model makes the data be physically implemented in the database. The description of how the building of data will take place is found in the physical model (Bentacourt, 2010).

Critical Review

            Everything when done in extreme becomes “poisonous”. The computer experts can develop particular software or control systems that will prevent the misuse of warehousing and data mining. Regulations and laws that govern the use of warehousing should be implemented and the offenders taken serious actions such as huge fines or the suspension of operating licenses.

References

Betancourt, L. (2010). How Companies are using your Social Media Data. Retrieved from http://www.enterpriseirregulars.com/5706/the-top-10-trends-for-2010-in-analytics-business-intelligence-and-performance-management/

Greenfield, L. (2005). The Case for Data Warehousing. Retrieved from http://www.dwinfocenter.org/casefor.html

Greenfield, L. (2005). The Case against Data Warehousing. LGI Systems Incorporated. Retrieved from http://www.dwinfocenter.org/against.html

Hadfield, M. (2009) Case Study: Jaeger uses data mining to reduce losses from crime and waste Retrieved from http://www.computerweekly.com/Articles/2009/02/23/234953/case-study-jaeger-uses-data-mining-to-reduce-losses-from-crime-and.htm

Thearling, K. (2009). An Introduction to Data Mining. LGI Systems Incorporated. Retrieved from http://www.thearling.com/dmintro/dmintro_2.htm