Abstract The

common belief is that a Bayesian network may achieve better performance with a more complex structure than a simpler structure which usually sacrifice the power of knowledge representation. It is often thought that having some additional nodes than necessary for knowledge representation does not, at least, deteriorate the performance of a Bayesian network. In practice, people often add some unnecessary nodes in order to make the network structure clearer to understand, or to reduce the dimensions of conditional probability tables of network nodes for the convenience of knowledge engineering. However, based on my experience in building Bayesian networks for web service diagnosis, such practice for convenience may cause unexpected results. Adding unnecessary nodes and growing the complexity of a Bayesian network may sometimes result in a worse performance. Such kind of practice should be taken care of very cautiously.