Naslov
Subgroup discovery for actionable knowledge generation: defiences of classification rule learning and lessons learned

Autori
Lavrač, Nada ; Gamberger, Dragan ; Flach, Peter

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Prooceedings of the ICML-2002 Workshop on Data Mining Lessons Learned / Lavrač, Nada; Motoda, Hiroshi; Fawcett, Tom - Sydney : University of New South Wales, 2002, 48-55

Skup
ICML-2002 Workshop on Data Mining Lessons Learned

Mjesto i datum
Sydney, Australija, 08.07.2002. - 12.07.2002

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
actionable knowledge; subgroup discovery

Sažetak
This paper discusses actionable knowledge generation. Actionable knowledge is explicit symbolic knowledge, typically presented in the form of rules, that allows the decision maker to recognize some important relations and to perform an action, such as targeting a direct marketing campaign, or planning a population screening campaign aimed at detecting individuals with high disease risk. The disadvantages of using standard classification rule learning for this task are discussed, and three subgroup discovery implementations are outlined. Case studies, a medical and two marketing ones, are used to present the lessons learned in solving problems requiring actionable knowledge generation.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA