Difference between revisions of "Test"

From Virtual Reality, Augmented Reality Wiki
Jump to: navigation, search
Line 1: Line 1:
 
==Introduction==
 
==Introduction==
−
A supervised learning classification task is an example of a machine-learning task. This type of problem requires that a model be trained on a labeled dataset. Each pair consists of an input object (such an image) as well as a label (such the object's classification). The model is then used to predict the label of unknown input objects.
+
A false negative (FN), in machine learning, is a type error that occurs when a model incorrectly classes a sample as negative even though it is actually positive. A false negative (FN) is when a model fails identify a positive sample.
  
−
==The Iris Dataset==
+
FNs are especially important in certain applications, such as medical diagnosis or fraud detection, where it is costly to fail to identify positive samples.
−
The [[Iris flowers data set]] is a well-known example of a classification issue. The data set includes 150 samples of iris flowers. Each sample has four features: the sepal width and length, and the length and width of the petals. The samples are from one of three classes: [[Iris vericolor]], [[Iris virginica]], and [[Iris setosa]. It is necessary to build a model from the data to predict the class of new Iris flowers based on their features.
 
  
−
==Support Vector Machine (SVM)==
+
==Examples==
−
Support Vector Machine]] algorithm. The SVM algorithm determines the best decision boundary (also called a hyperplane) to separate the different classes within the feature space. The decision boundary on which new input objects fall determines how they are classified.
+
In a medical diagnosis scenario, a FN could occur when a model that is supposed to identify patients with a specific disease incorrectly classifies a patient's health as healthy, when in fact they have the disease. This could result in the patient not receiving the treatment they need, which could have severe consequences.
  
−
==Example Implementation in Python==
+
A fraud detection scenario would see a FN if a model that is designed to detect fraudulent transactions incorrectly classifies a fraudulent deal as legitimate. This could result in financial loss as the fraudulent transaction is approved.
−
Here is an example of how one might implement the Iris classification task using the SVM algorithm in [[Python (programming language)|Python]]:
 
  
−
```python
+
==Measures==
−
from sklearn import datasets
+
There are many measures that can be used to assess the performance of a model, including the false positive rate (FNR), and the false discovery (FDR).
−
from sklearn import svm
 
  
−
# Load the iris dataset
+
FNR is the sum of the total number positive samples and the number of FNs. It is the percentage of positive samples incorrectly classified as being negative.
−
iris = datasets.load_iris()
 
−
X = iris.data
 
−
y = iris.target
 
  
−
# Create an SVM classifier with a linear kernel
+
The FDR is the sum of the number of FNs and the total number that were predicted to be positive. It is the percentage of negative samples that are incorrectly classified to be positive.
−
clf = svm.SVC(kernel='linear')
 
−
 
 
−
# Train the classifier on the iris dataset
 
−
clf.fit(X, y)
 
−
 
 
−
# Predict the class of a new, unknown iris
 
−
new_iris = [[5.0, 3.6, 1.3, 0.25]]
 
−
predicted_class = clf.predict(new_iris)
 
−
 
 
−
# Print the predicted class
 
−
print(predicted_class)
 
−
```
 
−
 
 
−
This code would output the class of the new iris, for example 'setosa'
 
  
 
==Explain Like I'm 5 (ELI5)==
 
==Explain Like I'm 5 (ELI5)==
−
Machine learning allows computers to learn without having to be programmed. One example is a game in which the computer attempts to identify what kind of flower it is seeing based on photos of other flowers it has seen. The computer is shown pictures of different flowers and is told what kind of flower it is looking at. After looking at many examples, the computer can see a new photo of a flower to determine what type it is.
+
False negatives are when a computer claims something is false when it is actually true. A false negative is when a computer claims that a person isn't sick, but in fact they are. This can lead to the person not receiving the medication they need.

Revision as of 09:49, 25 January 2023

Introduction

A false negative (FN), in machine learning, is a type error that occurs when a model incorrectly classes a sample as negative even though it is actually positive. A false negative (FN) is when a model fails identify a positive sample.

FNs are especially important in certain applications, such as medical diagnosis or fraud detection, where it is costly to fail to identify positive samples.

Examples

In a medical diagnosis scenario, a FN could occur when a model that is supposed to identify patients with a specific disease incorrectly classifies a patient's health as healthy, when in fact they have the disease. This could result in the patient not receiving the treatment they need, which could have severe consequences.

A fraud detection scenario would see a FN if a model that is designed to detect fraudulent transactions incorrectly classifies a fraudulent deal as legitimate. This could result in financial loss as the fraudulent transaction is approved.

Measures

There are many measures that can be used to assess the performance of a model, including the false positive rate (FNR), and the false discovery (FDR).

FNR is the sum of the total number positive samples and the number of FNs. It is the percentage of positive samples incorrectly classified as being negative.

The FDR is the sum of the number of FNs and the total number that were predicted to be positive. It is the percentage of negative samples that are incorrectly classified to be positive.

Explain Like I'm 5 (ELI5)

False negatives are when a computer claims something is false when it is actually true. A false negative is when a computer claims that a person isn't sick, but in fact they are. This can lead to the person not receiving the medication they need.


Retrieved from "https://xinreality.com/mediawiki/index.php?title=Test&oldid=25886"