Skip to main content

Advertisement

Table 10 Baseline and best results for NB, SVM, CRF

From: A comparison and user-based evaluation of models of textual information structure in the context of cancer risk assessment

   Acc. F-Measure
S1    OBJ METH RES CON        
  BL .29 .23 .23 .39 .18        
  NB .82 .85 .75 .85 .71        
  SVM .89 .90 .81 .90 .90        
  CRF .85 .87 .72 .87 .81        
S2    BKG OBJ METH RES CON REL FUT     
  BL .25 .13 .08 .22 .40 .13 - -     
  NB .76 .79 .25 .70 .83 .66 - -     
  SVM .90 .94 .88 .85 .91 .88 1.0 -     
  CRF .85 .92 .69 .77 .88 .75 - .33     
S3    HYP MOT BKG GOAL OBJT EXP MOD METH OBS RES CON
  BL .15 - .10 .06 .04 .06 .11 - .13 .24 .15 .17
  NB .53 - .56 - - - .30 - .32 .61 .59 .62
  SVM .81 - .82 .62 .62 .85 .70 - .89 .82 .86 .87
  CRF .71 - .74 .49 .72 .67 .59 - .58 .71 .56 .82