THE CURRENT MARKET CONDITIONS
 

 

 

 

 

 

 

 

 

 

 

 

 

For the appraisal example, the neural network has learned about historical patterns that allow it to predict the appraised value from descriptions of houses based on the contents of the training set. There is no guarantee that current market conditions match those of last week, last month, or 6 months ago—when the training set might have been made. New homes are bought and sold every day, creating and responding to market forces that are not present in the training set. A rise or drop in interest rates, or an increase in inflation, may rapidly change appraisal values. The problem of keeping a neural network model up to date is made more difficult by two fac­ tors. First, the model does not readily express itself in the form of rules, so it may not be obvious when it has grown stale. Second, when neural networks degrade, they tend to degrade gracefully making the reduction in perfor­ mance less obvious. In short, the model gradually expires and it is not always clear exactly when to update it.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The solution is to incorporate more recent data into the neural network. One way is to take the same neural network back to training mode and start feed­ ing it new values. This is a good approach if the network only needs to tweak results such as when the network is pretty close to being accurate, but you think you can improve its accuracy even more by giving it more recent exam­ ples. Another approach is to start over again by adding new examples into the training set (perhaps removing older examples) and training an entirely new network, perhaps even with a different topology (there is further discussion of network topologies later). This is appropriate when market conditions may have changed drastically and the patterns found in the original training set are no longer applicable. The virtuous cycle of data mining described in puts a premium on measuring the results from data mining activities. These measurements help in understanding how susceptible a given model is to aging and when a neural network model should be retrained. A neural network is only as good as the training set used to generate it. The model is static and must be explicitly updated by adding more recent examples into the training set and retraining the network (or training a new network) in order to keep it up-to-date and useful