M.L. Southcott and R.E. Bogner (1994). Classification of incomplete data using neural networks. Technical report University of Adelaide, Australia.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... Others have also sought to define a benchmark for machine learning [58, 62, 63, 66, 138] or connectionist [97] models and have remarked on the lack of scientific rigor in case studies [83, 96, 97] Many others have conducted case studies most of which suffer the limitations noted above [14, 43, 44, 53, 68, 87, 108, 119, 121, 122, 132, 137]. One exception to this rule is the Statlog project [88] conducted by many machine learning researchers from October 1990 to June 1993 under the ESPRIT program of the European community. With this notable exception, it is the scope and rigor (reproducibility) of this case study that distinguishes ....

M.L. Southcott and R.E. Bogner (1994). Classification of incomplete data using neural networks. Technical report University of Adelaide, Australia.

....attribute taking on the value unknown a situation which is clearly undesirable in a classifier. Finally, as each class is modeled by a single Gaussian or multinomial and the data points are assumed to be unlabeled, the Cheeseman et al. 1988) algorithm is in fact a form of soft clustering. Southcott and Bogner (1993) have approached the problem of classification of incomplete data using an approximation to EM for clustering. In the E step, the observed data are classified using the current mixture model, and each data point is assigned to its most likely class. The parameters of each class are then ....

Southcott, M. and Bogner, R. (1993). Classification of incomplete data using neural networks. In Proceedings of the Fourth Australian Conference on Neural Networks (ACNN '93), Sydney, Australia.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC