From Virtual Reality, Augmented Reality Wiki
|
|
| (37 intermediate revisions by the same user not shown) |
| Line 1: |
Line 1: |
| − | ===False Positive Rate (FPR) in Machine Learning== | + | <gallery mode="packed"> |
| − | The false positive rate (FPR), in machine learning, is a measure for the percentage of negative instances that are incorrectly categorized as positive. It is simply the sum of the number of falsely positive instances and the total number negative instances. FPR is often used with the true positive rate, also known as recall or sensitivity, to evaluate the performance and effectiveness of a binary classifier.
| + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image1.jpg |
| − | | + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image2.jpg |
| − | [[Sensitivity and Specificity|Sensitivity]] and [[Specificity (medical test)|specificity]] are two other measures of a binary classifier's performance that are related to the FPR. Specificity is the ratio of true negatives to all negatives, while Sensitivity measures the ratio of true positives to all positives.
| + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image3.jpg |
| − | | + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image4.jpg |
| − | In a [[Receiver operating characteristic|receiver operating characteristic (ROC) curve]], the FPR is plotted against the true positive rate. The ROC curve shows the trade-off between FPR and TPR at different threshold settings for a binary classifier. The ROC curve can be used to indicate the performance of the classifier in general. Each point represents a threshold setting.
| + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image5.jpg |
| − | | + | File:ZYBER_Charging_Dock_for_Meta_Quest_2_image6.jpg |
| − | ==Explain Like I'm 5 (ELI5)==
| + | </gallery> |
| − | False positive rate (FPR) can be used to determine how often a test which is supposed to give "no" actually gives "yes". False positives can be for example when a test claims you have a disease but you don't really have it. It is important to understand how often this happens in order to decide if the test should be used.
| |
Latest revision as of 16:47, 21 July 2023
