NASA Ames Res. Ctr. Intro. to IND Version 2.1, GA23-2475-02 edition, 1992.  Home/Search   Document Not in Database   Summary   Related Articles

....can be prohibitively expensive, especially if the cardinality of S is large. SLIQ uses a hybrid approach to overcome this issue. If the cardinality of S is less than a threshold, MAXSETSIZE, then all of the subsets of S are evaluated 3 . Otherwise, a greedy algorithm (initially proposed for IND [8]) is used to obtain the desired subset. The greedy algorithm starts with an empty subset S 0 and adds that one element of S to S 0 which gives the best split. The process is repeated until there is no improvement in the splits. This hybrid approach finds the optimal subset if S is small and ....

....tree as secondary metrics. The ideal goal for a classifier is to produce compact, accurate trees in a short time. 5.2 Experimental Setup The performance evaluation of SLIQ was divided into two parts. The first part compares SLIQ with the classifiers provided with the IND classifier package [8]. The IND package implements two of the most popular decision tree classifiers: CART [4] and C4 (a predecessor of C4.5 [10] These implementations are henceforth referred to as IND Cart and IND C4. Since the IND classifiers handle only datasets that fit in memory, the comparison used datasets ....

NASA Ames Res. Ctr. Intro. to IND Version 2.1, GA23-2475-02 edition, 1992.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC