Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

1) Classification performance is often measured with the following metrics: Accu

ID: 3597365 • Letter: 1

Question

1) Classification performance is often measured with the following metrics: Accuracy, Precision, Recall and F-measure. Match theses metrics to the application domain below (one for each only), where it would be most suited. Justify your answer with explanation. a) For balanced datasets, like for predicting the dataset to distinguish between cat and dog images. b) For imbalanced datasets, such as medical diagnosis data. c) For detecting suspicious samples from incoming network packets. d) For evaluating web search results.

Explanation / Answer

Answer: B

For imbalanced datasets, such as medical diagnosis data.

Explanation:

Countless examinations depend on grouping models. For example, a characterization model can be utilized to recognize potential tumor patients from their blood tests. Execution assessment of such model is basic to choose the best parameters and furthermore to contrast different models and a similar usefulness. The Receiver Operating Characteristics (ROC) plot has been routinely used to assess such characterization models.

The ROC plot was initially created by electrical architects amid World War II to recognize adversary objects from their radar signals. It has been utilized as a part of an extensive variety of fields including life sciences from that point forward. In light of its prevalence, the attributes of the plot have been all around considered. For example, one of the potential drawbacks of the ROC plot is that it can delude when connected to unequivocally imbalanced datasets.

Numerous datasets in life sciences are normally imbalanced. In any case, the ROC plot has been the most-broadly utilized assessment measure notwithstanding when the dataset is unequivocally imbalanced. Here, we uncover a few potential issues related with imbalanced datasets and furthermore demonstrate the upsides of the exactness review plot, which is an option assessment measure, over the ROC plot.