Skip to Main Content
Table 13 

Results for our three models trained on the Toefl11 data set and tested on EfCamDat, using continuous classifier outputs. Best result per column in bold, best result per row grouping is underlined.

 FeatureAccuracy (%)
Baselines Random Baseline 11.1 
Majority Class Baseline 11.1 
Single Vector Baseline 24.7 
Current Best (Malmasi and Dras 2015) 28.4 
  
Oracles Accuracy@2 42.0 
Accuracy@3 53.6 
Oracle 62.4 
  
Ensembles Plurality Voting 25.5 
Borda Count 26.9 
Mean Probability 28.4 
Median Probability 27.2 
  
Meta-classifier Linear SVM 31.9 
Logistic Regression 31.9 
Ridge Regression 31.4 
LDA 32.3 
  
Meta-classifier Linear SVM 32.0 
Bagging Logistic Regression 32.0 
Ridge Regression 31.8 
LDA 32.4 
 FeatureAccuracy (%)
Baselines Random Baseline 11.1 
Majority Class Baseline 11.1 
Single Vector Baseline 24.7 
Current Best (Malmasi and Dras 2015) 28.4 
  
Oracles Accuracy@2 42.0 
Accuracy@3 53.6 
Oracle 62.4 
  
Ensembles Plurality Voting 25.5 
Borda Count 26.9 
Mean Probability 28.4 
Median Probability 27.2 
  
Meta-classifier Linear SVM 31.9 
Logistic Regression 31.9 
Ridge Regression 31.4 
LDA 32.3 
  
Meta-classifier Linear SVM 32.0 
Bagging Logistic Regression 32.0 
Ridge Regression 31.8 
LDA 32.4 
Close Modal

or Create an Account

Close Modal
Close Modal