Knowledge Discovery in Databases

KNOWLEDGE DISCOVERY IN DATABASES 1

KnowledgeDiscovery in Databases

Alot of information is being collected due to the emergence ofelectronic systems and management tools which make it easier forhumans to analyze and keep records of their day to day activities.Software packages have aided in the knowledge discovery and datamining. (KDD) is the way in which aperson is able to establish meaningful and understand variouspatterns in data. Data mining is the process of obtaining patterns ormodes of data.

Thereare tools which enhance the entire development of KDD rather than thedata mining step. Tools which are in place today are the generictools which operate differently from the data source making more timeto be executed in data export and import and post processing(Zimmermann, 2006). From these tools there are various featureswhich include:

  1. Accessibility of data: the ability to access data source reduces the data transforming

  2. Offline /online data access: online data is where queries coordinate against databases while in offline analysis is performed with a snapshot.

  3. Underlying data model: tools input in form of a table where each record has a fixed number of attributes

  4. Query language: the language enables the user to process data and how to direct discovery process

Inthe KDD process there are data mining methods which are used forobtaining patterns from data. These methods ensure that differentgoals have been met. Data mining goals may be in the followingcategories:

  1. Data processing: there may be selection, filtering and transformation depending on the goals.

  2. Regression: it’s the analyzing of values upon values of other attributes.

  3. Prediction: there is prediction of value for a particular item

Clustering:this is grouping of clusters with similar characteristics together.

Variousmethods have different goals. They can be classified into:

  1. Case –based reasoning: it’s whereby there is solving of a problem by use of past experiences and solutions.

  2. Neural networks: these are systems modeled after human brains which contain networks that work similar to neurons in human brains.

Decisiontree: it’s where a node stands in place of a test or decision onthe data item

Inconclusion, there is acquisition of useful information fromdatabases. In the society today, various firms gather and puttogether databases for they realize the importance of them in makingbusiness decisions. However the application of KDD is a long process.Nevertheless, demands by the public bring out the availability oftools and users are trained on how to use them. This makes adevelopment of the KDD technology.

References

Zimmermann,H. J. (2006). Knowledge management, knowledge discovery, and dynamicintelligent data mining. Cyberneticsand Systems: An International Journal,37(6),509-531.