Questions tagged [auc]

The tag has no usage guidance.

14 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
2
votes
0answers
25 views

Fast PR / ROC curves and corespondings AUPR / AUROC

I find myself in a position of calculating numerous PR / ROC curves and their associated area under the PR curves (AUPR) / area under the ROC curve (AUROC). Its is quite easy to perform those ...
2
votes
0answers
209 views

AUC on ROC Curve near 1.0 for Multi-Class CNN but Precision/Recall are not perfect?

I am building a ROC Curve and calculating AUC for multi-class classification on the CIFAR-10 dataset using a CNN. My overall Accuracy is ~ 90% and my precision and recall are as follows: ...
2
votes
0answers
15 views

Connection between prob output LogisticReg/SVM and ROC

I have the following ROC generated using LPOCV and Logistic regression or SVM (l2 norm). Now, let's say I have a test set containing 10 patients and I get that the probabilities of those patients to ...
1
vote
0answers
40 views

Best practice to select precision vs. recall threshold for a binary classifier

I have a logistic regression model in Scikit-Learn doing a binary classification. Looking at the Roc curve for the classifier I can see that it performs really well: The AUC score is 0.99 which is ...
1
vote
0answers
15 views

How to compute a confidence interval for AUC?

I found that in results of several binary classification problems, people report an AUC value together with a CI. I wonder how those CIs are computed. Is there a close-formed formula to compute them ...
1
vote
2answers
45 views

Will oversampling help with generalization (small imbalanced dataset)?

I have an imbalanced dataset (2:1 ratio) with about 60 patients and 80 features. I performed Recursive Feature Elimination (RFE) and stratified cross validation to reduce the features to 15 and I get ...
1
vote
0answers
223 views

How to create an roc plot and calculate AUC for an svm (that does not return probabilities)?

I have some SVM classifier outputting final classifications for every sample in the test set, something like 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1 and so on. The "...
1
vote
0answers
107 views

Regression for binary classification and AUC metric

In the kaggle forums I found an example model where someone was using XGBRegressor for a binary (0/1) classification problem (sorry, cannot find the link any more). This was for a competition where ...
1
vote
1answer
487 views

XGBOOST missing_value feature degrades my performance?

I'm training an xgboost model for gout disease on a training set I sampled 1-to-7 case-control ratio (enriched in cases). I have 220 features and I reach a cross-...
0
votes
0answers
8 views

Model performance in different snapshots varying

I am trying to solve this problem. A medical representative needs to visit some doctors' clinics and for that a model will generate probability scores for visiting a clinic. I ma using a tree based ...
0
votes
0answers
8 views

LightGBM model improvement when the focus is on probability prediction

I am building a binary classifier using LightGBM. The goal is not to predict the outcome as such, but rather to predict the probability of the target even. To be more specific, it's more about ranking ...
0
votes
0answers
20 views

When to use AUROC OvR vs. AUROC OvO?

For multiclass problems, there are 2 versions of the AUROC metric: the AUROC OvR and AUROC OvO. Does anyone know in what particular cases we would use AUROC OvR vs. AUROC OvO? In the general academic ...
0
votes
0answers
27 views

Trying to plot REC curve and calculate AUC in R but getting error

I am working on regression machine learning techniques and calculating AUC (plotting REC) by using below paper link: http://homepages.rpi.edu/~bennek/papers/rec.pdf My Code; REC <- function(y_val , ...
0
votes
0answers
13 views

What is a right way to compare AUCROC between logreg and logregs on segments?

I built a logistic regression on the entire sample (first model), as well as regressions on separate pieces (disjoint) of same sample (set of models). What is the best way to compare the effectiveness ...