Maral Aghazi – 500287851
Professor Roger De Peiza
"As we and our students write messages, post on walls, send tweets, upload photos, share videos, and “like” various items online, we’re leaving identity trails composed of millions of bits of disparate data that corporations, in the name of targeted advertising and personalization, are using to track our every move” (McKee, 2011). Data mining has become extremely prevalent in today’s society given the vast advancements in technology, particularly the internet. Companies as well as the government are able to discover mass amounts of information about people, which those people are unaware of themselves. This allows companies to research and cater to its consumers in quick and efficient ways, by sifting through large amounts of data and compiling the information necessary for their success. In terms of meeting customers’ needs by evaluating trends and group behaviours it is beneficial to society, however there are also detrimental consequences given the sum of personal content available on many social networking platforms. This paper will define data mining, the impact it has grown to have and it’s further development in the future. Data mining is a concept with which most of us may not be familiar in terms of its prevalence and importance. Data mining is defined as an “analysis of large pools of data to find patterns and rules that can be used to guide decision making and predict future behaviour” (Laudon, Laudon & Brabston, 2011). This can be used to discover trends for essentially everything. While purchasing our daily cup of Starbucks coffee in the morning may seem meaningless and irrelevant to us, we could very well be part of a compilation of data used for further research. There are many ways in which information can be obtained via data mining. The first of these methods is association. Association refers to the relation between the occurrence of one event and that of another (Laudon, Laudon & Brabston, 2011). An example of this would be, a study done for purchasing patterns at a supermarket, which found that 65 percent of the time people who purchased corn chips would purchase cola as well. When there is a promotion on corn chips, cola is purchased 85 percent of the time (Laudon, Laudon & Brabston, 2011). Sequences are events that are linked over time. For example when someone purchases a car, they will most probably buy winter tires when needed. Classification recognizes current patterns of a group by using rules concluded by existing items. An example would be cell phone service providers studying customer behaviours in order to formulate an appropriate plan to maintain their relationships with loyal customers. Clustering is much like classification, however the groups have not yet been formed. Lastly there is forecasting, which uses existing data to predict future data. These processes were introduced during the second-generation of data mining in 1995 (He, 2009). Data mining is occurring all around and most people remain completely unaware. “This data collected from different sources if processes properly, can provide immense hidden knowledge…” (Dharminder, 2011). As previously mentioned, the recording of transactions in supermarket in order to discover customer patterns is just one example. Another would be Facebook. Through Facebook, simply “liking” an activity, interest, band, etc., gives advertising agencies the ability to use this information to post ads that adhere to that certain user. These ads then pop up on the side of that particular users news feed, which not only caters to that users interests, but also creates efficient and effective promotion for many products and services. Facebook calls this form of data mining “instant personalization” (McKee, 2011). Data mining has been around since the 1960’s, but information has not always been as easy to collect and store as it is...