Which data mining algorithm to use?

Out of so many data mining algorithms in SQL Server 2005, it is important to be able to select the right algorithm that meets the business needs. In addition, depending on the algorithm used, it is also important to understand the nature of the data that needs to be used as input. Remember - "Garbage In - Garbage Out"...

Decision Trees
- Predict whether a consumer will buy a particular product
- Finding groups of common items
- Determining whether a customer would likely churn or switch
- Generally, to predict the value of a discrete attribute

Naive Bayes/Neural Network
- Predict whether a consumer will buy a particular product
- Generally, to predict the value of a discrete attribute

Time Series
- Forecasting (e.g. predicting future revenue, sales volume)

Association Rules
- Finding groups of common items in transactions
- Deriving rules (with different confidence) that if customer buy product A, he/she is most likely to buy product B.

Sequence Clustering
- Clickstream analysis of customer web-site behaviour
- Predicting a customer next possible step in a sequence of actions.

Clustering
- Segmentation of customer based on demographic data