Thakshashila: /* SEO Keywords */

2025-06-10T06:25:11Z

SEO Keywords

← Older revision		Revision as of 06:25, 10 June 2025
Line 67:		Line 67:

	specificity in machine learning, true negative rate, sensitivity vs specificity, specificity formula, confusion matrix specificity, model evaluation metrics, binary classification		specificity in machine learning, true negative rate, sensitivity vs specificity, specificity formula, confusion matrix specificity, model evaluation metrics, binary classification
			[[Category:Artificial Intelligence]]

Thakshashila: /* Specificity vs Sensitivity */

2025-06-10T05:22:18Z

Specificity vs Sensitivity

← Older revision		Revision as of 05:22, 10 June 2025
Line 40:		Line 40:
	== Specificity vs Sensitivity ==		== Specificity vs Sensitivity ==

	These are complementary metrics:		These are [[complementary metrics]]:

	* '''Sensitivity''' = Ability to detect positives		* '''Sensitivity''' = Ability to detect positives

Thakshashila: Created page with "= Specificity = '''Specificity''', also known as the '''True Negative Rate (TNR)''', is a performance metric in binary classification tasks. It measures the proportion of actual negative instances that are correctly identified by the model. == Definition == : $\text{Specificity} = \frac{TN}{TN + FP}$ Where: * '''TN''' = True Negatives – actual negatives correctly predicted * '''FP''' = False Positives – actual negatives incorrectly predicted as positi..."

2025-06-10T05:21:29Z

Created page with "= Specificity = '''Specificity''', also known as the '''True Negative Rate (TNR)''', is a performance metric in binary classification tasks. It measures the proportion of actual negative instances that are correctly identified by the model. == Definition == :<math> \text{Specificity} = \frac{TN}{TN + FP} </math> Where: * '''TN''' = True Negatives – actual negatives correctly predicted * '''FP''' = False Positives – actual negatives incorrectly predicted as positi..."

New page

= Specificity =

'''Specificity''', also known as the '''True Negative Rate (TNR)''', is a performance metric in binary classification tasks. It measures the proportion of actual negative instances that are correctly identified by the model.

== Definition ==

:<math> \text{Specificity} = \frac{TN}{TN + FP} </math>

Where:
* '''TN''' = True Negatives – actual negatives correctly predicted
* '''FP''' = False Positives – actual negatives incorrectly predicted as positives

Specificity answers the question: '''"Out of all real negative cases, how many did the model correctly classify as negative?"'''

== Alternate Names ==
* True Negative Rate (TNR)
* Selectivity

== Simple Example ==

Suppose a test is used to detect a rare disease. Out of 1,000 healthy people:

* 950 are correctly identified as healthy → TN = 950
* 50 are incorrectly diagnosed with the disease → FP = 50

:<math> \text{Specificity} = \frac{950}{950 + 50} = \frac{950}{1000} = 0.95 = 95\% </math>

This means the test correctly identifies 95% of healthy people.

== Importance of Specificity ==

Specificity is vital when false positives can cause unnecessary stress, cost, or risk.

=== Real-World Scenarios ===

* Medical Testing: Avoiding false diagnoses of a disease (e.g., not telling a healthy person they are sick).
* Spam Filters: Ensuring genuine emails are not classified as spam.
* Fraud Detection: Not labeling legitimate transactions as fraudulent.

== Specificity vs Sensitivity ==

These are **complementary metrics**:

* '''Sensitivity''' = Ability to detect positives
* '''Specificity''' = Ability to rule out negatives

Together, they form a balanced evaluation of a model, especially in medical or safety-critical applications.

== Combined Evaluation: ROC Curve ==

Receiver Operating Characteristic (ROC) curves plot:

* Sensitivity (True Positive Rate) vs.
* 1 − Specificity (False Positive Rate)

This helps visualize the trade-off between catching positives and avoiding false alarms.

== Related Metrics ==

* [[Sensitivity]] – True positive rate
* [[Precision]] – Positive prediction correctness
* [[Recall]] – Same as Sensitivity
* [[F1 Score]] – Harmonic mean of Precision and Recall
* [[Confusion Matrix]] – Base for all metrics

== SEO Keywords ==

specificity in machine learning, true negative rate, sensitivity vs specificity, specificity formula, confusion matrix specificity, model evaluation metrics, binary classification

Specificity - Revision history

Thakshashila: /* SEO Keywords */

Thakshashila: /* Specificity vs Sensitivity */