Abstract. The ability to restructure a decision tree efficiently enables a variety of approaches to decision tree induction that would otherwise be prohibitively expensive. Two such approaches are described here, one being incremental tree induction (ITI), and the other being non-incremental tree induction using a measure of tree quality instead of test quality (DMTI). These approaches and several variants offer new computational and classifier characteristics that lend themselves to particular applications.
Ada is a registered trademark of the U.S. Government, Ada Joint Program Office. Seutes b UNCLASSIFIED IDA Log No. NOD 87-32518 ... , DEFINITONS GAPM p beS biOMul bame 10 iPW INe i of No f &Lum NII uo so ONd ammlmollam moraf Iedl umdidp Mia A polimm. They smmli l immul II mowa polsm which Ia) have a dked aul ddcisins ulmha MaOm Ppma or (b) adiru hums of alIfeag Is Fit Ezmm f bank, of Cumsauw oe powM, or to) ad*uamam Sow" no w h iaveI I uumumi Iapilumu. IDA Ruuduif m rsulm by odadu pomals of qpu -is m mur lbigij me mob of spuusalymhs MWMl ipma Io pbue d a bus, ad beaoMM pmols %um met Palm m imuiuwud Wbumes *A "e mdI masim abar to hm .oeaud NbMlusP pi In p~udsaft* ul Memohmdum Report. MA Muououadu Rmpmrim amgued fo go ummelmmm d ft spaiemime d aile mmmi mbebimldaue I 11l dweads. bma d maimI luaidu lablma"upe adMlbelto imb mallabl. polkela" and relm se lls of alpm or d e~n p"p and paid awmsle hwI lumollu *A Is am mile mmul amd onma volsarle bmuksona reid Nieru mollmgu, m kdumaid drededevlwimpdn -Nome d so lei bulglol. Realm of gmm~adm Ropulm Is mimi to Sin ~mos aid blowili mus. The ammaieMNWA ,k u nnmasemy by kdIm nil Idnm mmmammd is spuuu se sbi deallomime bymeh pumei. Ass Mapill. It a ap ill Wae dmandm mm mmdmtd mimi mmmba MIA 90 C O31 for me OIdMpeha d 00191m0. IIS pm&imaof.fN hIDA damMINmiW demalimale IM deiuemina by In Dspubeud d Omm, aM ubudi VIS IMmmlub be ONNNuu IS Iae. me mili posillso d Su spay.I [Tisa pp kms beass ralWA by MA b smm 0 N musoi blO aldmada Neig m d ugeaall. ad NOmi amyeltt ItMem dm m t mNISII m mSM IOm go m LAIs
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.