Purpose of Data Mining
What is Data Mining?
Data mining is extracting patterns, trends or information from data sets.
How Data is Collected
Almost everything you do online can be tracked and used to data analytical purposes. Some online methods of data mining include:
Search engine queries - recording of recent internet searches
Social media activity - voluntary information given when creating a profile (name, email, sex, interests, friends)
Internet Tracking Software
Online purchasing history
There are also offline ways to collect information, including:
Credit card transactions
Loyalty cards - Eg Flybuys, Woolworths Rewards, Myer One
Information handed for surveys or competitions
The main purpose of data mining is for marketing purposes to which this information is collected to a build a profile about a user, with personal information including their name, age, interests, sex, interests, dislikes and geographical location.
The idea is that once marketers know who you are, they can better promote products and services based on your interests and personal attributes. Marketers use algorithms to detect and identify certain patterns in online activity to build a customer profile.
How Target Saw the Teen Pregnancy Before Dad Did
In the United States, Target sent a teenage girl personalised vouchers for baby related items, such as nappies and baby formula. The father was outraged, believing that Target was suggesting her daughter was pregnant. Target apologised but later discovered that the teenage girl was in fact, pregnant.
Target figured it out by analysing the shopping habits of the teenage girl by linking a Guest ID number to a credit card, name, email address and other personal information. The Guest ID tracked that the girl had been buying pregnancy related items and used a pregnancy prediction score, to guess when the baby was due. Based on this information, Target US sent the teenage girl personalised vouchers for baby products, with the analysis of her shopping records used to determine she was in fact pregnant and the baby was due soon.
Watch Kirsten Drysdale from the ABC show 'The Checkout' explain the role of data mining in today's consumeristic society.
Data Mining | The Checkout | ABC 1
Data mining can be used to link symptoms together to identify an illness. For example, if you have a running nose and high temperature you could be diagnosed as having a fever.
Web MD uses voluntary inputted data from its online forms to determine possible illnesses.
Businesses can use data mining by analysing the activities of its employees and detecting any unusual activity. For example, suspicions would be raised if a junior employee accessed sensitive files, used only by senior employees.
In a wider context, data mining can be used to detect suspicious activity such as the National Security Hotline which relies on data sent voluntary by members of the public.