Here are the basic technical details for assembling the knowledge base for your Expert System:

1. Identify the events that are relevant to your application. You need to focus on input or causes and outputs or effects. Outputs can be instructions, outcomes, consequences, or conclusions. For example, one outcome of a litigation case is losing an appeal to a large bank. A possible cause is a biased judicial system that gives significant systematic advantages to big banks over small business owners. Such bias can be quantified and can affect the outcome of the litigation -- if you don't do something about it.
2. Assign a probability to each input event or cause. If you do not know the exact probability of the event, you can assign a range of probabilities. When you do so, we can conduct what is called a sensitivity analysis. Such analysis shows you how the probabilities of the output events vary with changes in the probabilities of the input events.
3. Relate each output or effect to its possible inputs or causes. For this, you need to specify the if-then rules of your decision making process.
4. Specify the degree to which you believe the rules to be valid or certain. Some rules are absolutely certain. Others, may be approximate rules of thumb. You can use uncertain rules, so long as you can quantify the degree to which they are uncertain. We'll show you how to do this in the next section.

Once you complete the above steps, you have the KNOWLEDGE BASE of your Expert System.