Data Mining and Model Simplicity : A Case Study in DiagnosisGregory

We describe the results of performing data mining on a challenging medical diagnosis domain, acute abdominal pain. This domain is well known to be diicult, yielding little more than 60% pre-dictive accuracy for most human and machine di-agnosticians. Moreover, many researchers argue that one of the simplest approaches, the naive Bayesian classiier, is optimal. By comparing the performance of the naive Bayesian classiier to its more general cousin, the Bayesian network clas-siier, and to selective Bayesian classiiers with just 10% of the total attributes, we show that the simplest models perform at least as well as the more complex models. We argue that simple models like the selective naive Bayesian classiier will perform as well as more complicated models for similarly complex domains with relatively small data sets, thereby calling into question the extra expense necessary to induce more complex models.