Data Mining in E-Commerce
This is an excerpt from the paper...
Data mining as applied to e-commerce is a breakthrough technology that can gather information in an automated fashion and build models used to predict customer purchasing decisions with remarkable accuracy. Through data mining, e-tailers can customize their offerings and pricesùpersonalizing them to each online customerùand achieve much higher sales. Simply by knowing what type of customer is likely to purchase a particular item, e-tailers can accurately target their customers and obtain add-on sales that rapidly add up to big dollars. Although e-tailers used to be at a disadvantage in comparison with merchants in bricks-and-mortar stores, data mining has evened the playing field, allowing e-tailers to achieve the equivalent of a relationship with the customer without ever seeing his or her face. Since data mining largely occurs behind the scenes, unbeknownst to consumers, it has been criticized for privacy concerns, but most consumers appreciate the convenience of having items suggested to them that are exactly what they like. Within the software and databases that support data mining, a number of techniques are used to model and analyze the data, from simple association or clustering to more advanced logistic regression. Many of these are used in conjunction with one another to provide more precise profiling of customers and their purchasing preferences. Data mining has the potential to revolutionize e-commerce, reducing many of its
. . .
t a ôverification-based approachö where the user forms a hypothesis about data interrelationships and then uses the tools to verify whether the hypothesis is correct or incorrect, relying on ôthe intuition of the analyst to pose the original question and refine the analysis based on the results of potentially complex queries against a databaseö (Moxon). The validity of the results is limited by the analystÆs ability to pose appropriate questions and ôthink out of the boxö (Moxon). Data mining, however, starts with the data rather than the analyst, using ôdiscovery-based approachesö such as pattern-matching and other algorithms to identify the key relationships in the data (Moxon).
Association
Association is one of the simplest data mining techniques, using a ômarket-basket analysisö that treats the purchase of a number of items as one transaction and looks for trends across a multitude of transactions to identify natural buying patterns (Moxon). Association approaches usually employe ôconfidence-rated rulesö that express the results in statements such as ô80 percent of all transactions in which beer was purchased also included potato chipsö (Moxon).
Sequence-Based Analysis
Sequence-based analysis is a more comprehensive app
. . .
Some common words found in the essay are:
Advantages E-Commerce, Analysis Factor, Logistic Regression, Detection CHAID, Abstract Data, Miningö Data, Sandoval Essentially, Internet Internet, Moxon Finally, Fourth July, data mining, , modeling techniques, multiple modeling techniques, retrieved december, using multiple, using multiple modeling, multiple modeling, value using, factor analysis, value using multiple, december 8 2005, retrieved december 8, techniques dataö, 8 2005,
Approximate Word count = 3339
Approximate Pages = 13 (250 words per page)
More Essays on Data Mining in E-Commerce
|
Customer Service
to Over
32,000 Professionally Written Papers!!!
