Accuracy Calculation using confusionmat?

Kindly I have an issue with accuracy calculation, please could you help ?

y_train = [ 1 1 1 4 4 3 3 5 5 5 ]; % for x_train

%x_test : has no true labels.

predictedLabel=[ 1 2 3 4 5 ]; %predicted lables for x_test

I used the code below :

group=y_train ; % 10
grouphat=predictedLabel; % for test 5 test data
C=confusionmat(group,grouphat);
Accuracy = sum ( diag (C)) / sum (C (:)) ×100;

but get error:

Error using confusionmat (line 75)

G and GHAT need to have same number of rows

I had error since the test data is more or less than the train ( no lables)?

No true label for test data ( semi supervised learning )

Thank you

Answer 1

[Spruha]

Hi Shawin,

I see you are encountering an error while using ‘confusionmat’. This issue arises because MATLAB’s ‘confusionmat’ function requires both input vectors—the true labels and the predicted labels—to have the same length and to correspond to the same set of data samples. Since you do not have the true labels for your test set, it is not possible to use ‘confusionmat’ to evaluate your results in this case.

Please refer to documentation of ‘confusionmat’ for more information: https://in.mathworks.com/help/stats/confusionmat.html

Instead, you can consider using internal validation metrics such as the silhouette score or the Davies-Bouldin index.

Please refer to below documentation links to know about silhouette score and Davies-Bouldin index:

• https://in.mathworks.com/help/stats/clustering.evaluation.silhouetteevaluation.html
• https://in.mathworks.com/help/stats/clustering.evaluation.daviesbouldinevaluation.html

Hope this helps!