Accuracy

What is the definition of accuracy in a confusion matrix?

What is the formula for accuracy?

We are calculating the accuracy with a confusion matrix in a classification problem. Which element (True Positive, False Positive, True Negative, and False Negative) we do not need for doing so?

When should accuracy be used as a performance metric for classification models?

Accuracy will yield misleading results if the dataset is:

Which option is another formula of accuracy?

You got a set of results about the performance of a classification model, which are precision = 0.8, recall = 0.73, and specificity = 0.78. What is the overall accuracy of the model?

You are evaluating a model on a result of face mask detection using computer vision algorithms for a bus station. The information you got are precision = 70% and recall = 80%. What is the answer if you want to know the general accuracy of this case?

A confusion matrix (see the table below) was created by a Data Scientist to evaluate the performance of the classifier of a smoke detector test. What is the accuracy of the classifier?

A grocery store would like to improve their business by analyzing their customers. In fact, they are using Computer Vision algorithms to detect the gender of their customers. According to the confusion matrix below, can we evaluate the model performance with the accuracy and what is the accuracy in this case?

Seattle-Tacoma International Airport (SEA) is using a Computer Vision application to detect whether passengers wear face masks or not. You as a Data Scientist, need to evaluate the performance of the classifier. Based on the confusion matrix below, can we evaluate it with the accuracy and what is the accuracy in this case?

Amazon is training their Alexa devices (voice assistants) with a new wake word (e.g., Alexa!). During testing, 300 sentences were spoken to simulate a living room environment where people were talking. “Alexa!” was mentioned ten times in those 300 sentences to see if the voice assistant would wake up, and the wake word was not mentioned in the remaining  sentences (positive case is when a wake word is mentioned). According to the result of the confusion matrix below, is accuracy the best metric to evaluate the classifier’s performance? What is the accuracy in this case?

Google is training their Google Home voice assistants with a new wake word (e.g., Ok Google!). During testing, 100 sentences were spoken to simulate a dining room environment where people were talking. “Ok Google!” was mentioned 50 times in those 100 sentences to see if the voice assistant would wake up, and the wake word was not mentioned in the remaining 50 sentences (positive case is when a wake word is mentioned). Based on the result of the confusion matrix below, is accuracy a good metric to evaluate the classifier’s performance? What is the accuracy in this case?