Analytics In Data Mining

Submitted By salita24
Words: 1568
Pages: 7

Abstract
Simply collecting data for research is nearly a faux pas in today’s competitive web-market. Analysts are now looking toward the predictive analytics of association discovery in web and data mining, to find Business Intelligence of clustering sub=populations while eliminating errors to keep collected data valid. In the midst this data crunch are fears of lost privacy. Do not fear. Creative innovations are bringing mash-ups to our diversity.

Data Analytics Report

Useful information, knowledge and finding some unexpected results can “strike it rich” with added creative thinking. Data mining supplies analysts, investors, and traders with customers buying patterns, historical trading rules, even fraudulent behavior for insurance claims. Predictive analytics is used in web mining by analyzing user’s movements from one web content to another. Collecting the data of where a user browses and the content they are seeking can become knowledge if the analyst understands the patterns (Turban & Volonino, 2011).

An Association Discovery Algorithm is a tool of data mining where new rules are discovered such that if one item is present then another will also be found. This type of knowledge benefits analyst’s predictability of future probabilities and is very useful to the marketing department, (Ranjan, 2008).

A traditional example you may have heard about association discovery shows when diapers are purchased so is beer, (Guo, 2002). This is good to know for several reasons. The marketers can plan to always have enough beer to cover when diapers sell. Doctors and insurance planners may recommend a different product to replace alcohol for a young family’s health.

It once was business intelligence when the web masters counted on-line transaction processing. That was, however, a rudimentary measurement of success when the dot com era began. Analysis must be made to determine directions for improvement. Web mining has developed into the analytics of web usage, content searches, and Web-metrics such as how many visitors, how many page views and reference sites, (Panian, 2012). This new age data mining has given great insight for the development of smart-phones.

If you have a product selling so-so on line or anyway in the market, clustering data mining can help identify how unique your customers are. Clustering refers to a subpopulation of people or things within a larger group, like Chinatown or a certain type of customer. When you understand who exactly your customers are then you can focus advertisements on that specific group, (Singh, 2012).

The term “noisy information sources” refers to levels of inaccuracy that common data most likely contains. Data mining may not be error free if the algorithms used to acquire results assume the source data is error free. This is the reasoning behind using algorithms that recognize one way or another, that the data sources may contain errors for a number of reasons. When possible data errors are removed using an algorithm that data is then considered “cleansed”, (Wu & Zhu, 2008).

Mr. Falkvinge considers the debate whether the threat to data privacy is from the government or corporations as pointless because both are bad for us. He claims: Target knows when a teenage girl is pregnant before her parents. Visa can predict a divorce one year ahead. Google and governments have near mind control powers and high tech wiretapping abilities, (Falkvinge, 2012).

Mr. Wolford, following Facebook’s promise to provide better analytics for marketers was surprised how fast the Center for Digital Democracy sent a letter to the federal Trade Commission demanding an investigation to determine if Facebook has violated their recently adopted Consent Order, (Wolford, 2012).

Mr. Timberg of the Washington Post warns, the use of Google services will likely include being data mined. Google recently consolidated its privacy policy after confronting legal problems in Europe. SafeGov.Org is