Testing Random Forests for Unix and Windows

0370539 - ÚI 2012 RIV CZ eng L4 - Software
Jiřina, Marcel - Jiřina jr., M.
Testing Random Forests for Unix and Windows.
Internal code: Friendly random forest ; 2011
Technical parameters: Modifikovatelný zdrojový kód ve Fortranu 95 a výkonné (exe) moduly pro Windows a Linux, návody k použití a zkušební příklady - viz. Jiřina, Marcel ; Jiřina jr., M. Testing Random Forests for Unix and Windows. Prague : ICS AS CR, 2010. 33 s. (TR V-1075).
Economic parameters: Podstatné zlepšení použitelnosti metody Random Forests tím, že byla odstraněna nutnost zasahovat do zdrojových textů. Modernizace přenesením z Fortranu 77 do Fortranu 95.
R&D Projects: GA MŠMT(CZ) 1M0567
Institutional research plan: CEZ:AV0Z10300504
Keywords : random forest * RandFor * multivariate data * clustering * classification * data mining * probability estimation
Subject RIV: JC - Computer Hardware ; Software

The Random Forest is a method and also a program for data clustering and classification. Especially in classification the Random Forest method appears to be the best approach perhaps among all others up to now. In our modification in Fortran 90 there are binaries for Windows and Linux and information for different tasks is passed with help of arguments. After a brief description of our modification of RandForest the detailed manual follows. The next Chapter describes testing data corpora used and gives results of the Random Forest program.
Permanent Link: http://hdl.handle.net/11104/0204323