This preview shows page 1. Sign up to view the full content.
Unformatted text preview: r of physical joins, and
provides simple metadata.
On the other hand, in the snowflake schema model, the data warehouse consists of
normalized dimension tables, by attribute level with each smaller dimension table
pointing to an appropriate aggregated fact table. Each dimension table has one key
for each level of dimension's hierarchy. The lowest key, for example, the store key
of store dimension will join store fact table, which will also have product and time
keys. An example of a snowflake schema is shown in Figure 16.18. A practical
way of designing a snowflake schema is to start with a star schema and create
snowflakes with queries. The main advantage of snowflake schema is that it
provides very good performance for queries involving aggregation. DATA MINING
The term 'mining' literally means the operations involved in digging for hidden
treasures. Similarly, 'data mining' is used for the operations involved in digging
out critical information from within an organization’s stored data for better
decision support. It is the process of identifying interesting and useful patterns
from large databases or data warehouses. As shown in Figure 16.14, data mining is
the core part of the KDD process.
Formally, data mining is defined as "the nontrivial process of extracting implicit,
previously unknown, and potentially useful information from data". The meaning
of the key terms used in this definition is as follows:
1. Nontrivial process. The term 'process' here means that data mining involves
many steps including data preparation, search for patterns, knowledge evaluation,
and refinement - all repeated in multiple iterations. This process is 'nontrivial' in
the sense that it goes beyond simple search for quantities, and involves search for
structure, models, and patterns.
2. Implicit information. This means that the information extracted as the result of
data mining is from within an organization's stored data.
3. Previously unknown information. This means that the structures, models, o...
View Full Document
- Spring '14