{"id":613,"date":"2012-12-19T12:46:19","date_gmt":"2012-12-19T17:46:19","guid":{"rendered":"http:\/\/henry.olders.ca\/wordpress\/?p=613"},"modified":"2018-03-10T18:35:31","modified_gmt":"2018-03-10T23:35:31","slug":"hammingnn-classifier-results-with-various-datasets","status":"publish","type":"post","link":"https:\/\/henry.olders.ca\/wordpress\/?p=613","title":{"rendered":"HammingNN classifier: results with various datasets"},"content":{"rendered":"

2012-12-19<\/p>\n

This table presents results of using the HammingNN classifier with a variety of publicly available datasets that are frequently used for development and testing of classifier paradigms. For comparison purposes, I\u2019ve included results from 3 other classifier paradigms.<\/p>\n

Note: you may need to widen your browser window to view all the columns.<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Data Set<\/b><\/td>\nCases<\/b><\/td>\nAttributes<\/b><\/td>\nClasses<\/b><\/td>\nHammingNN
\nleave-one-out<\/b><\/td>\n		HammingNN
\n10-fold<\/b><\/td>\n		Parameters<\/b><\/td>\nXCSTS<\/b><\/td>\nC4.5\u00a0<\/b><\/td>\nNaive Bayes<\/b><\/td>\n<\/tr>\n
Anneal<\/td>\n898<\/td>\n32D 6C<\/td>\n5<\/td>\n99.67%<\/td>\n99.496%<\/td>\nn=21 i=28 k=2<\/td>\n97.7%<\/td>\n98.6<\/td>\n86.6<\/td>\n<\/tr>\n
Breast cancer Wisconsin<\/td>\n699<\/td>\n9D<\/td>\n2<\/td>\n96.567%<\/td>\n96.383%<\/td>\ni=8 j=3<\/td>\n95.9<\/td>\n94.5<\/td>\n96.0<\/td>\n<\/tr>\n
BUPA<\/td>\n345<\/td>\n6C<\/td>\n2<\/td>\n70.72%<\/td>\n69.67%<\/td>\nn=15<\/td>\n67.1<\/td>\n65.0<\/td>\n54.8<\/td>\n<\/tr>\n
Glass<\/td>\n214<\/td>\n9C<\/td>\n7<\/td>\n76.64%<\/td>\n75.45%<\/td>\ng n=19 i=5<\/td>\n71.8<\/td>\n67.4<\/td>\n48.1<\/td>\n<\/tr>\n
Ionosphere<\/td>\n351<\/td>\n32C<\/td>\n2<\/td>\n93.447%<\/td>\n92.829%<\/td>\nn=12 i=23<\/td>\n90.1<\/td>\n90.0<\/td>\n82.5<\/td>\n<\/tr>\n
Iris<\/td>\n150<\/td>\n4C<\/td>\n3<\/td>\n98.000%<\/td>\n97.960%<\/td>\ni=2 n=3<\/td>\n94.7<\/td>\n94.8<\/td>\n95.5<\/td>\n<\/tr>\n
Leukemia<\/td>\n72<\/td>\n5147C<\/td>\n2<\/td>\n97.222%<\/td>\n95.028%<\/td>\nn=8 i=11<\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Mushroom<\/td>\n8124<\/td>\n22D<\/td>\n2<\/td>\n100%<\/td>\n100%<\/td>\ni=7<\/td>\n100.0<\/td>\n100.0<\/td>\n95.8<\/td>\n<\/tr>\n
Prostata<\/td>\n102<\/td>\n12533C<\/td>\n2<\/td>\n94.118%<\/td>\n92.451%<\/td>\nn=6 i=3<\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Soybean<\/td>\n683<\/td>\n35D<\/td>\n19<\/td>\n94.290%<\/td>\n93.634%<\/td>\nz i=25<\/td>\n85.1<\/td>\n91.9<\/td>\n92.8<\/td>\n<\/tr>\n
Vehicle<\/td>\n846<\/td>\n18C<\/td>\n4<\/td>\n70.567%<\/td>\n70.051%<\/td>\ni=18 n=13<\/td>\n74.1<\/td>\n72.4<\/td>\n45.1<\/td>\n<\/tr>\n
Vowel<\/td>\n990<\/td>\n10C<\/td>\n11<\/td>\n98.687%<\/td>\n98.087%<\/td>\ng n=13<\/td>\n66.0<\/td>\n80.1<\/td>\n63.2<\/td>\n<\/tr>\n
Wine<\/td>\n178<\/td>\n13C<\/td>\n3<\/td>\n97.191%<\/td>\n96.708%<\/td>\nn=5<\/td>\n95.6<\/td>\n93.3<\/td>\n97.2<\/td>\n<\/tr>\n
Yeast<\/td>\n1484<\/td>\n8C<\/td>\n10<\/td>\n58.154%<\/td>\n57.023%<\/td>\nu=5 i=6<\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Zoo<\/td>\n101<\/td>\n16D<\/td>\n7<\/td>\n98.02%<\/td>\n97.911%<\/td>\ni=12<\/td>\n95.1<\/td>\n93.0<\/td>\n95.7<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Data Sets: from the UIC Machine Learning Repository; I used the versions of these datasets downloaded from the Orange website: http:\/\/orange.biolab.si\/datasets.psp<\/a><\/p>\n

Attributes: D = discrete (ie nominal or ordinal or categorical); C = continuous-valued.<\/p>\n

Parameters:<\/p>\n