Business Intelligence & Analytics

It shows a hierarchy of components contributing to an organizations intelligence.

•Data comes in the form of raw observations, e.g. temperature, scores, and bits of news.

•Information adds context to data; it is created by analyzing relationships and connections between the data.

•Knowledge is created by using the information for action. Knowledge answers the question "how".

•Wisdom is created through the use of knowledge and through reflection. It is the ultimate level of understanding.

Condensation: Data is summarized. Unnecessary depth is eliminated.

Calculate: Data is analyzed mathematically or statistically.

Contextualization: We know why the data was collected.

Correction: Errors removed.

Categorization: Unit of analysis is known.

Comparison: How does this situation compare to others?

Consequence: What implications does the information have for decisions and actions?

Connection: How does this bit of information relate to others?

Conversation: What do other people think about this information?

A Data Warehouse is a subject-oriented, integrated time-variant, non-volatile collection of data in support of management's decision-making process.

top-down: user requirements

bottom-up: which data sources are available

Extract data from source systems

Clean up data and transform it

Consolidate (index the data, summarize)

Restructure keys

Load data into the DW

Maintain the metadata

Refresh the DW with updated data

ETL tasks ensure that the data in the DW is.

Relevant

Useful

Quality

Accurate

Accessible

•Multipart keys: record key structures with built-in meaning

•Multiple local standards

•Complexity of integrating Multiple files

•Missing values

•Duplicate values

•Element names: variation in naming conventions in fields from source to source.

•Element meanings: different interpretations of the name of an element.

•Inconsistent Input formats

•Inconsistent data values

•Referential integrity constraints

•Name and address