DyClee-C: a clustering algorithm for categorical data based diagnosis
Résumé
In data-based diagnostic applications, large amounts of data are often available but the data re- mains unlabelled because labelling would require too much time and imply prohibitive costs. The different situations, e.g. normal or faulty, must hence be learned from the data.
Clustering methods, also qualified as unsupervised classification methods, can then be used to create groups of samples according to some sim- ilarity criterion. The different groups can sup- posedly be associated to different situations. Nu- merous algorithms have been developed in recent years for clustering numeric data but these meth- ods are not applicable to categorical data. How- ever, in many application domains, categorical features are key to properly describe the differ- ent situations. This paper presents DyClee-C, an extension of the numeric feature based DyClee al- gorithm to categorical data. DyClee-C is applied to two data sets: a soybean data set to diagnose the disease soybean plants and a breast cancer data set to assess the current diagnosis in terms of recur- rence events and prognose possible relapse.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...