.        .        .        .        .        .        .        .        .        .                Data Mining in Business ...
..........Data Mining in Business Research"Data mining is not a single technique, but a set of statistical techniques that...
Although there is a tendency to think of data mining purely as a recentphenomenon, the core of its process has many simila...
..........    Data mining has had a long history as an organizational analysis tool prior to thewidespread use of computer...
Web site log data mining recently began to converge with (CRM) CustomerRelationship Management Software (Berry). CRM typic...
..........    As relationships with online retailers deepen as in the case with 12 year oldAmazon.com, the emergence of ex...
"When the data mining process is not well understood, all the clever techniques andalgorithms get applied to the wrong dat...
.           .           .           .           .           .           .           .           .           .BIBLIOGRAPHYI...
Malhotra, Y. (2007). Scientific method, and evolution of scientific thought. Retrieved on11 April 2008 from http://www.bri...
Upcoming SlideShare
Loading in …5
×

Data Mining in Business Research

446 views

Published on

Data Mining in Business Research

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
446
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
4
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Data Mining in Business Research

  1. 1. . . . . . . . . . . Data Mining in Business Research. . . . . . . . . . Edgardo Donovan TUI University BUS 504 – Dr. Roger Rensvold Module 3 – Case Analysis Monday, February 18, 2008
  2. 2. ..........Data Mining in Business Research"Data mining is not a single technique, but a set of statistical techniques that is usedto identify trends, patterns, and relationships in data. The increasing amount ofavailable data on the web is driving renewed interest in data mining, which isgaining in popularity." (Allen) Data mining as we understand it today is the evolving practice of tabulating largequantities of information and using analysis tools to discern patterns that can offerinsight on how an organization can better adapt its operational model. Although datamining has been practiced in various forms using a variety of business machines overthe past century, it has become more widespread with the advent of computers andhas become a very popular method for improving online business practices throughthe analysis of web site statistics logs. 2
  3. 3. Although there is a tendency to think of data mining purely as a recentphenomenon, the core of its process has many similarities with past human forms ofinquiry. Examples of this can be found with the similarities between post positivistresearch methodology and data mining. Both deal with analyzing information andcategorizing data into different groups of variables. Both disciplines are intent ongathering a series of quantitative and qualitative data that will form the basis of a planof action or a hypothesis. In certain data mining projects a technique calledregression is utilized whereby formulas are generated in an attempt to form the basisof a hypthesis (Chappie). In both methodologies the bulk of the data patterns areused to scrutinize and ultimately serve as a basis to prove or disprove a hypothesis.Similarly to academic research, data miners begin a project with the goal ofanswering a series of business related questions. Typical inquiries may involvefinding out how many customers purchased certain combinations of products orduring what days they were the most likely to shop. However, the encountered datamay not be conducive towards answering preconceived questions. It may not beuncommon, either through regression or deductive analysis, that unexpected logicalpatterns are found thereby expanding the scope and focus of the overall project"Human beings depend so heavily on patterns in their day-to-day lives that they tendto see patterns even when they don’t exist. If you look at the night-time sky, youprobably do not see a random arrangement of stars, but rather, the Big Dipper, orthe Southern Cross, or Orion’s Belt. Some of you even see astrological patterns andportents that can be used to predict the future. This was an early form of datamining!" (Berry) 3
  4. 4. .......... Data mining has had a long history as an organizational analysis tool prior to thewidespread use of computers. Similarly to today, it was used extensively byAmerican political parties throughout the 19th century as a way to develop state andlocal level demographic profiles so as to best understand where to focus politicalresources. Gerrymandering, otherwise know as the practice of delineating politicaldistricts according to historical political voting tendencies, relied heavily upon ondeducing patterns and trends from localized voting records. Albeit in a much more tedious and slower fashion, businesses have alwaysattempted to elicit logical trends and patterns from operational and financial recordsin the hope of being better able to predict how to adapt sales, marketing, andoperational management techniques to the ever changing competitive marketplace.Some common methods of data mining have revolved around individuating peak salepropensity periods, optimal product bundling periods, and the timing of promotionalsales. Data mining in its modern form involves the utilization of software to gather,categorize, and statistically analyze large quantities of scanned data. It has taken onincreased importance with the advent of the Internet. The use of data mining vis-à-visweb site log analysis has grown exponentially over the past 15 years and is expectedto continue to do so in the future (Allen). 4
  5. 5. Web site log data mining recently began to converge with (CRM) CustomerRelationship Management Software (Berry). CRM typically involves maintainingextensive customer databases while creating a suite of automated and semi automatedcorrespondence events that attempt to offer product and service promotions that areaffinity customized and timed for each customer."WebTrends Visitor Intelligence allows you to act on deep insight into the personbehind the visitor to drive microtargeted marketing online experiences that increaseengagement and build loyalty!" (Berry) Since its inception in the mid 1990s, Log Analyzer by Web Trends, Inc. hasarguably been the leading web analysis software in its class enabling webmasters andmarketing departments to compile reports regarding their web site audience visitpatterns. The information that can be gleamed off of web site logs is limited to visitdynamics and cannot extract identifying demographic information due to userprivacy laws. However, these reports are very useful because at a minimum they caninform companies which web sites are referring visitors, the size of their webaudience, their most popular pages, their most downloaded files, their busiest times,and their audience computer operating system of choice. With the advent of the latestapplications combining both log analyzers and CRM software, marketers are able toautomate some of the customer interaction like never before thereby expanding theiropportunities to increase sales. 5
  6. 6. .......... As relationships with online retailers deepen as in the case with 12 year oldAmazon.com, the emergence of extensive proprietary online affinity marketingcustomer databases in the marketplace is steadily increasing. For example, each timeI visit the Amzon.com web site I personally get customized offers based on a varietyof topics related to past purchases which have been painstakingly tracked over theyears. Data mining also has applications online in the form of contextualized webrobots. Robots are a term to describe simple applications that project themselvesthroughout the Internet with the goal of locating, categorizing, downloading, andreformatting files of a specific content type. This methodology can be useful in thedevelopment of resource materials or general research. In 1999 a company by thename of Flipdog.com made millions of dollars by implementing a job search spiderthat was able to recognize all the job postings on the Internet while organizing themthrough a centralized real-time web database interface. Data mining whether it involves web site logs, affinity marketing customerdatabases, contextualized web spiders, or operations management processimprovement has limited applications. Although it involves, finding, gathering,categorizing, and analyzing large quantities of data, the responsibility of ensuring thatsensible hypotheses are derived from the latter falls on individual common sense. 6
  7. 7. "When the data mining process is not well understood, all the clever techniques andalgorithms get applied to the wrong data, in the wrong ways, and yield wrong results.A corollary is that the skills of the human data miner and that individual’s knowledgeand intuition, about how to coax meaning from recalcitrant data, are more importantthan tools and techniques." (Berry) Data mining as we understand it today is the evolving practice of tabulating largequantities of information and using analysis tools to discern patterns that can offerinsight on how an organization can better adapt its operational model. Although datamining has been practiced in various forms using a variety of business machines overthe past century, it has become more widespread with the advent of computers andhas become a very popular method for improving online business practices throughthe analysis of web site statistics logs. 7
  8. 8. . . . . . . . . . .BIBLIOGRAPHYI. Works Cited Allen, Cliff. (1999). Mining for gold. Retrieved on 11 April 2008 from http://retailindustry.about.com/gi/dynamic/offsite.htm?site=http%3A%2F%2Fclickz.com %2Farticle%2Fcz.681.html Berry, Michael. (2008). Understanding variables. Retrieved on 11 April 2008 from http://www.crm2day.com/library/EpAZpuAFyAzpbJTtYp.php Chappie, Mike. (2008). Data mining: an introduction. Retrieved on 11 April 2008 from http://databases.about.com/od/datamining/a/datamining.htm Webtrends, Inc. (2008). Identify, target, and engage your customers like never before. Retrieved on 11 April 2008 from https://www.webtrends.com/upload/DS_WebTrends_Visitor_Intelligence.pdfII. Works Consulted Allen, Cliff. (1999). Mining for gold. Retrieved on 11 April 2008 from http://retailindustry.about.com/gi/dynamic/offsite.htm?site=http%3A%2F%2Fclickz.com %2Farticle%2Fcz.681.html Berry, Michael. (2008). Understanding variables. Retrieved on 11 April 2008 from http://www.crm2day.com/library/EpAZpuAFyAzpbJTtYp.php Chappie, Mike. (2008). Data mining: an introduction. Retrieved on 11 April 2008 from http://databases.about.com/od/datamining/a/datamining.htm Cook, T.D. (1983). Quasi-Experimentation: Its Ontology, Epistemology, and Methodology. In Gareth Morgan (Ed.), Beyond Method. Dereshiwsky, M. (1998). Understanding hypotheses. Northern Arizona University Dereshiwsky, M. (1998). Understanding variables. Northern Arizona University Hutchinson, Paul. (2007). Discussion on the process of doing research. Retrieved on 11 April 2008 from http://www.angelfire.com/biz/rumsby/ARES.html Krishnamurthy, S. (n.d.). An empirical study of the causal antecedents of customer confidence in e-tailers. First Monday Journal 8
  9. 9. Malhotra, Y. (2007). Scientific method, and evolution of scientific thought. Retrieved on11 April 2008 from http://www.brint.com/papers/science.htm.Murray, David (1998). Group-randomized trials. Oxford University Press.Ozz, E. (2001). Organizational commitment and ethical behavior: An empirical study ofinformation system professionals. Journal of Business EthicsNational Academy of Sciences. (2007). On being a scientist. Retrieved on 11 April 2008from http://www.nap.edu/readingroom/books/obas/.Oulton, Tony. (1995). Management research for information. Library Management.Trochim, W. (2007). Research methods website. Retrieved on 11 April 2008 fromhttp://www.socialresearchmethods.net/index.htm.Webtrends, Inc. (2008). Identify, target, and engage your customers like never before.Retrieved on 11 April 2008 fromhttps://www.webtrends.com/upload/DS_WebTrends_Visitor_Intelligence.pdf 9

×