Tagging - ML

              Sometimes classification model won’t work the best on some data. For instance, in the image (Figure 1), we can see multiple animals in the picture. Our classification model won’t be able to classify this input correctly because there are multiple classes (animals) on it. We want to give our model option to say that this image contains a cat, dog, rooster, and a donkey.

                                                              Figure 1 - multi-labeled classification

              The problem of learning to predict classes that are not mutually exclusive is called multi-label classification. Auto-tagging problems are typically best described as multi-label classification problems.

              Nikola Andric