ABSTRACT
Like Expert Choice, Strategizer rates alternative policies, but it does so in a way that the user finds impossible to control. This is because it uses a simulated, neural network (McClelland and Rumelhart 1988; Nelson and Illingworth, 1991). The latter is better at accommodating the non-linearities, discontinuities, inaccuracies and fragmented data (Noorderhaven, 1995) that are associated with policymaking. As such, Strategizer’s neural network may constitute a more effective method than traditional, statistical approaches for anticipating people’s policy choices. Moreover, it would theoretically get better and better at such anticipation the more it is used-the system would ‘self improve’, even when confronted with policies that it has not seen before. In short, Strategizer is potentially the quintessential software for assisting humans in the ‘anticipate’ phase of the policymaking process (Harrison and St John, 1994).
