GROUP PRESENTATION BY:
DATA BASE MANAGEMENT SYSTEM
A complete system used for managing digital
databases that allow storage of data, maintenance
of data and searching data.
Also known as Knowledge discovery in
Data mining consists of techniques to find out
hidden pattern or unknown information within a
large amount of raw data.
An example to make it more clear:
A grocery chain used the data mining capacity
of oracle software to analyze local buying patterns. They
found that when men bought any item on Thursday and
Saturday, they also tended to buy beer. They further
acknowledged that most of these customers did their
weekly grocery shopping on Saturdays instead of
Thursdays. Hence the grocery chain could use this newly
discovered info in various ways to increase revenue.
Data is being produced at a phenomenal rate and so the the amount of data stored has grown.
People expect more prominent information rather then huge number of raw data lying in databases.
Data mining provides relation between large data, which is a more useful information.
DATA MINING vs DATA-BASE TECHNIQUE
Data Mining activities differs from Database interrogation. Data mining identifies hidden pattern within data while
Database inquiries ask for retrieval of data.
Data mining is practiced on static data collection, called ‘DATA WAREHOUSE’, rather than ‘online’ databases which keep on updating.
FORMS OF DATA MINING
1) CLASS DESCRIPTION:
Class description deals with identifying properties that characterize a given group of data items, whereas class discrimination deals with identifying properties that divide two groups.
-For example, class description techniques would be used to identify characteristics of people who buy small economical vehicles. By identifying their salaries
2) CLASS DISCRIMINATION:
Class discrimination deals with identifying properties that divide two groups. Class discrimination are also the techniques that would be used to find properties that distinguish customers who shop for used cars from those who shop for new ones.
Cluster analysis tries to find properties of data items that lead to the discovery of groupings. For example, in analyzing information about people’s ages who have viewed a particular motion picture, cluster analysis might find that the customer.
It involves looking for links between data groups. It is association analysis that might reveal that customers who buy potato chips also buy beer and soda or that people who shop during the traditional weekday work hours also draw retirement benefits.
Outlier analysis is another form of data mining. It tries to identify data entries that do not comply with the norm. Outlier analysis can be used to identify errors in data collections, to identify credit card theft by detecting sudden deviations from a customer’s normal purchase patterns, and perhaps to identify potential terrorists by recognizing unusual behavior.
Tries to identify patterns of behavior over time. For example, sequential pattern analysis might reveal trends in economic systems such as equity markets or in environmental systems such as climate conditions
The Konstanz Information Miner, is an open source data
analytics, reporting and integration platform. Incorporates
components for machine learning and data mining.
KNIME is being used in :
-CRM customer data analysis,
-Financial data analysis.
Please join StudyMode to read the full document