Konferenzbeitrag
Encapsulated models for reasoning and decision support
Lade...
Volltext URI
Dokumententyp
Text/Conference Paper
Dateien
Zusatzinformation
Datum
2013
Autor:innen
Zeitschriftentitel
ISSN der Zeitschrift
Bandtitel
Verlag
Gesellschaft für Informatik e.V.
Zusammenfassung
In this paper we show how knowledge from the medical domain can be incorporated in classification in a way that improves the transparency of classification (the 'why'), and makes the classification less dependent on both the particular data set used for training and on peculiarities of the classification algorithm. We compare a decision tree incorporating a domain model with a tree built directly from the same data.