-[__Decision Tree and RandomForest__](#decision-tree-and-randomforest)
@@ -282,11 +283,6 @@ In order to get an idea about the performance of our baseline algorithms, that n
| **Precision (macro)** | 0.22 | 0.22 | 0.2 |
| **Precision (weighted)** | 0.28 | 0.22 | 1 |
For the evaluation process we used the following metrics.<br>
-**Accuracy:** The score that can be seen above is calculated by the score method in the sklearn.dummy.DummyClassifier module and shows the mean accuracy on the given test data and goldlabels.
-**Precision/Recall:**
-**F1-Measure:**
## __Classification Task__
After all the work we put into preparing our data and creating baseline classifiers it is now time to try out if our data can be classified with DecisionTree, RandomForest, NaiveBayes, SVM Kernel and Multilayer Perceptron classification algorithms.<br>
@@ -425,8 +421,19 @@ As our data is obviously not linearly separable we expected the __SVM Kernel__ a
## __Evaluation__
### __Accuracies__
### __Confusion Matrix__
The highest accuracy have the SVMs with features most important.This can be explained, among other things, by the fact that SVMs require few features. Too many features usually have a negative effect on the value of the machine prediction.
The results are also skewed by a most frequent bias. This is evident from the confusion matrix: for categories 2 to 5 (i.e., all categories except those for lowest alcohol consumption), there are almost no false positives. A large proportion of individuals actually report consuming very little alcohol (1); however, quite a few individuals are also incorrectly assigned to the lowest category.
Translated with www.DeepL.com/Translator (free version)
The following plots show the __learning curves__ of our classifiers in the first row. For this the training score is plotted compared to the amount of training data used in the process. The training score then can be compared to the cross-validation curve, which shows the testing scores in relation to the amount of data. The second row represents the amount of time used to train with the amount of training samples and in the third row shows how much time was required in training to get a certain score.
