| Popis: |
Abstract: Data mining has been instrumental in the extraction of some useful knowledge from data. The purpose of data mining has always been to focus on searching for methods to extract instead of revealing the implicit models of the data. The outcome of the process of data mining is the knowledge that is represented by different visualization techniques. Knowledge obtained through data mining is not effective without the intervention of a domain expert who uses that knowledge to make a decision. On the other hand, human participation has the potential to influence and predispose decisions. Human participation in the process of data mining is still subjective and cannot be automated. A possibility to look into this quandary is the conversion of these subjective factors into some measurable parameters. This predicament leads to the development of an area that can be referred to as "Wisdom Mining," which will consist of procedures to add wisdom to the extracted knowledge. Wisdom mining, if it is proposed as an extension to data mining, exhibits the need for certain factors, methods, and measures beyond the methods and measures used in the data mining process. The factors proposed in this article for a seamless transition from data to wisdom mining are context, utility, time, and location. There are two possibilities to use these factors for the extraction of wisdom from data. One is to develop new algorithms for wisdom mining from scratch, keeping these four factors as major placeholders. The second approach is to add these four factors to the existing algorithms of data mining to get wise patterns as outcomes. The paper proposed a second approach for laying the foundation of this new domain of wisdom mining. |
