Data Mining Techniques - Essay Example

- Predictive: using known observations to predict future events
- Forensic Analysis: uncharacteristic or odd data aspects are processed to extract patterns

There are three types of data mining activities performed by its users in an organization. The activities are:
- Episodic Mining: looks at data from a particular incident or an event such as a direct marketing campaign. This way, the data is understood or used for any future marketing campaigns. This type of technique is usually used by many analysts.
- Strategic Mining: it uses a larger set of data of an organization to gain overall recognition of particular measures e.g. profitability, liquidity, etc.
- Continuous Mining: Understanding of factors that affect things globally within a time frame and the entire factors that lead to such changes are known.
The three main drawbacks of using data mining are:
- Issues relating to the security of the information obtained by companies about individual customers. This information is prone to crimes such as hacking creating major security concerns for both customers and companies.
- Information obtained from various customers can be used in unethical ways. This information can be used in issues such as discrimination, etc.
- Many customers do not give out their information easily because of privacy complaints. This way it is not possible to accurately draw relationships or trends in any particular data as the data may be considered incomplete.
Airline companies use data mining techniques to improve their service to customers. The data that is mined to find hidden patterns is about frequent fliers. The issues selected for mining in the frequent flier program are first to identify customer characteristics, their most frequent flying zone, class, and period of the year in which they usually fly. Secondly, to discover patterns between the sectors based on the activities of the customers. The data mining technique is applied to customer data in three ways: category type, booking type, and sector type. Category type revealed that discretionary and invitees were having lower than average use usage whereas members flew more. Booking type disclosed that about 88% of the customers booked their tickets via agents.

Data mining can be used by many different industries. One type of industry is the supermarket industry. Data mining can help them identify customer buying patterns i.e. to place products in such a way that can lead to an increase in the buying behavior of the customer e.g. Wal-Mart’s case of nappies and beer. The other industry that can use data mining is the automobile industry. Customer preferences can be found and all future automobiles can be based on aspects usually preferred by customers (e.g. low CO2 emission cars and increased promotion of hybrid cars to control pollution). Read More