Pregled bibliografske jedinice broj: 81256
An Integrated Intelligent System for Data Mining and Decision Making
An Integrated Intelligent System for Data Mining and Decision Making // Proceedings of the 2002 Interbational Conference on Intelligent Engineering Systems / Lovrenčić, Alen ; Rudas, Imre A. (ur.).
Zagreb: Fakultet organizacije i informatike Sveučilišta u Zagrebu, 2002. str. 287-292 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 81256 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
An Integrated Intelligent System for Data Mining and Decision Making
Autori
Ujević, Filip ; Bogunović, Nikola
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the 2002 Interbational Conference on Intelligent Engineering Systems
/ Lovrenčić, Alen ; Rudas, Imre A. - Zagreb : Fakultet organizacije i informatike Sveučilišta u Zagrebu, 2002, 287-292
Skup
INES 2002, 6th Interbational Conference on Intelligent Engineering Systems
Mjesto i datum
Opatija, Hrvatska, 26.05.2002. - 28.05.2002
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
Data mining; Knowledge discovery; Decision making
Sažetak
A widespread proliferation of databases has created demand for powerful tools for turning stored data into useful knowledge. Analytical and predictive capabilities of data mining tools form the basis of business decision support systems. The paper explicates a novel data classification environment designed for real-world business problems that exhibit a high degree of undeterministic behavior. The environment is implemented in Java and uses SQL queries for direct database access. It employs a variation of Naive Bayes algorithm augmented with attribute selection for eliminating redundant attributes. Target concept is described using C4.5 decision tree generator. An embedded visual performance evaluation tool enables significantly faster convergence towards the final model. The predictive power of the classification environment is illustrated on a typical business problem.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb,
Institut "Ruđer Bošković", Zagreb
Profili:
Nikola Bogunović
(autor)