K15-2003 performance slightly below the flat classification . None of the other features
W10-0217 experiments and , for comparison , the flat classification results . A comparison of the
P08-3012 comparable to other state-of-the-art flat classification systems . Future work includes
W10-0217 is comparing hierarchical and flat classification , assuming all the other parameters
J14-4007 the hierarchy entirely , using a flat classification of just twelve sense categories
D13-1117 . This is intractable even for flat classification . Following Bishop ( 1995 ) and
D09-1118 better decision detection than a " flat classification " approach with a single " decision-related
C02-1155 flat category model , traditional flat classification is applied directly to the derived
C02-1155 the classification accuracy of flat classification on WebKB and DI data sets . Here
W10-0217 results in both experiments with flat classification shows that in both cases the
C02-1155 is reduced . As a consequence , flat classification may face with the sparseness
P08-3012 performance of state-of-the-art flat classification systems . 2 Related Works Most
W10-0217 global results are higher than flat classification , but the accuracy of the global
W08-0125 accuracy achieved by systems based on flat classifications . For the task of detecting decision
D09-1118 systems which use hierarchical vs. flat classification ) , and Section 6.3.3 presents
W10-0217 classes which used to dominate in flat classification but they no longer dominate in
W04-3201 support vector machines ( SVMs ) in flat classification tasks like text categorization
S13-1003 , we compared the traditional flat classification approach and our proposed hierarchical
D11-1095 clustering has focussed on acquiring flat classifications from corpus data , although many
C02-1155 multi-dimensional classification outperforms flat classification in most cases but sometime the
hide detail