More and more unstructured data is becoming available over time.

Suppose you buy something online at Amazon or Bol.com.

When you but something a certain event is triggered in the backend information system where the raw data is stored.

When this data is used in a management report, we can call it information.

It is possible that some companies see certain data as information where other companies see it as knowledge.

Predictive analytics: predict future behavior based on past behavior (Time series, clustering, decision trees, NN) 4.

Prescriptive analytics: make decisions or recommendation to achieve the best performance (simulation, optimization, AI).

Informationknowledge that could relate to: - Understanding customer preferences - Coping with competition - Identifying with opportunities - Enhancing internal efficiency BI solutions support the BI process by utilizing BI tools.

For example: a customer buys something, than a certain event is created in the sales database.

All the different databases have a local view on the business.

After that, a large set of techniques is available to provide insights for the decision maker.

So, its not comparable with multiple excel tables, because database tables are related to each other.

A database is also an environment to Manipulate data using a query language (e.g.

SQL), and it is often an clientserver architecture.

The clients are active (low processing and very thin) and the database server is relatively passive (but does a lot of processing, is thick).

It consists of 1 or more attributes that determine other attributes.

When a primary key contains multiple values, it is called a composite primary key.

Keys role is based on determinations: A B, C, D o If you know A, you can lookup B, C, D o B, C, D are functionally dependent on A Textual representation of the database tables.

Therefore you need to clean and convert the data.

Historical data: Next to that, the problem is time-related.

Mostly, databases in the backend do not have much historical data.

Data consolidation: Lastly, you want to aggregate the data.

Backward selection: start with a large set of variables and keep deleting variables that harm your model.

The next time I visit the site I will not get this popup anymore.

Data Mining For Business Analytics