Questions tagged [unbalanced-classes]

Data organized into discrete categories or *classes* may present problems for certain analyses if the number of observations ($n$) belonging to each class is not constant across classes. Classes with unequal $n$ are *unbalanced*.

Filter by
Sorted by
Tagged with
0
votes
0answers
9 views

GAMM input: can it have different lengths?

I am building a GAMM model on the difference between two types of utterances. F1 refers to the first formant of the utterance, which is a continuous variable. Utterance refers to the two different ...
0
votes
2answers
36 views

Is it ok a threshold of 0?

I am dealing with a classification problem with a dataset containing 60k rows: 69k are negative class, and 1k is positive. I trained my models and I obtained the confusion matrices with a threshold of ...
3
votes
1answer
24 views

correcting for extremely downsampled data: keras class_weight is hurting my model

I have an extremely imbalanced dataset (millions of times more negatives) for a binary classification NN model. I am aggressively downsampling solely for the purpose of making training time manageable,...
1
vote
1answer
20 views

How can I adjust predicted probabilities after resampling?

I have a real-world problem with severe imbalanced classes. I was able to get a good AUC and balanced accuracy after the implementation of a resampling technique. Now I want to "walk over the ROC&...
0
votes
0answers
10 views

roc auc for small class imbalance

I have a classification problem with class imbalance(1:6). I'd like to know if roc_auc is a valid metric for this level of imbalance. I know it's not good for severe imbalance, but what about a case ...
0
votes
0answers
8 views

Factorial Mixed ANOVA Unequal Gender Ratio [duplicate]

For my thesis I'm doing a factorial Mixed ANOVA in which I look at gender differences. As it's a mixed ANOVA, everyone goes through the same conditions. The problem however is that I have 10 men and ...
0
votes
0answers
3 views

ROSE (Random Over Sampling Examples) in python

I am currently working on imbalanced data topic. And I found a function in R called ROSE (paper). I understand from a high level how the function works, ...
0
votes
0answers
33 views

Dealing with classes in an imbalanced dataset

I have a dataset of continuous features and 4 classes. The classes counts are 1793, 246, 103 and 102. Adding data is quite difficult now. I've done classification with a random forest on the entire ...
0
votes
1answer
18 views

Precision and Recall for highly-imbalanced data

I have an imbalanced data with binary label where there are only 4% positive labels among all examples. I want to evaluate my model on the dataset, and I wonder what is the best way (best metric) to ...
0
votes
0answers
23 views

Low precision after smote

I am working on a classification problem with class imbalance. I implemented every kind of balance technique and I always get high accuracy, recall and roc (0.85) and low precision( around 0.50). Also ...
0
votes
0answers
10 views

Interpretation of confusion matrix after use of classe_weight in Logistic Regression

I am running a test to verify the effect of the unbalanced classes in binary classification. I am using logistic regression with L1 penalty for classifying (I optimised the penalty through CV and ...
0
votes
0answers
15 views

What does it tell me when my ensemble learners (classifier trees) are just as good with only one split?

Basically new to decision trees and ensemble classifiers. Looking for guidance based on what I am seeing. My goal really isn't to use a decision tree. It is to do a simple binary (A/B) ...
0
votes
1answer
38 views

Should I balance my dataset for binary classification?

I am running through a binary classification problem. My target variable is unbalanced the 1s are the 13% of the whole dataset and 0s 87%. Total number of observation 697, number of features 709. ...
3
votes
1answer
29 views

Is there an issue using an imbalanced covariate (not dependent) in logistic regression?

I am investigating data from a randomised control trial, where treatment allocation is done on a 2:1 ratio (2 patients on the experimental treatment for every 1 patient on placebo). 400 on ...
1
vote
1answer
26 views

A question about a logistic regression classifier performance (with and without resampling)

I am working on a dataset with 20 independent variables and 41188 instances. The task is a binary classification where the target variable has 36548 number of no's and 4640 of yes's. I have used ...
0
votes
0answers
10 views

How to analyze data from unbalanced fractional factorial design

I ran a 2x2x3 mixed fractional factorial design that collected reaction times for detecting targets across depths. Within-subject IVs: Cue depth (levels: near, mid, far) Depth validity (levels: ...
0
votes
0answers
22 views

Calibrating probability thresholds based on ROC curve for multiclass classification

I have built a network for the classification of three classes. The network consists of a CNN followed by two fully-connected layers. The CNN consists of convolutional layers, followed by batch ...
1
vote
0answers
43 views

Resample Unbalanced sequence data in deep learning doesn't have good effect?

I'm working on text classification using a deep learning approach. Because the data I use has unbalanced conditions, I try to implement data balancing techniques using the imblearn library. However, ...
0
votes
0answers
15 views

Is the PR AUC invariant under label flip?

The ROC-AUC curve is invariant under a flip of the labels. I don't know if it's a famous result, so I will give the proof below. My question is if the PR-AUC curve also has this property. I have not ...
0
votes
0answers
5 views

if i want well calibrated probabilities but have class imbalance what metric?

i am having some issues on trying to get a correct metric for an imbalanced problem. it is a credit risk problem where i am trying to predict default of a company so i care about probability output. i ...
0
votes
0answers
13 views

Dropout in highly unbalanced longitudinal data (WGEE)

I have found a lot of software and examples that uses Weighted Generalized Estimating Equations to deal with missing data in a balanced data set (equal time points). However, I have a very high ...
0
votes
0answers
26 views

High Validation F1 score but low testing F1 score

I am working on a dataset related to an insurance company and the objective is to predict if the insurance buyer will claim their travel insurance or not. Training data: https://raw.githubusercontent....
1
vote
1answer
21 views

Which metric to use to evaluate highly imbalance classification model performance

I have to do classification model to predict the possibilities of person getting cancer based on certain attributes. The data is highly imbalanced. As per client requirement I have to report model ...
1
vote
0answers
17 views

Binary classification, imbalanced dataset optimization: AUC vs logloss

I'm running optimization on an imbalanced dataset and need to define my optimization metric. I'm working on disease detection so maximizing AUC might not be the best solution, as the certainty of the ...
0
votes
0answers
17 views

The effect of an imbalanced dataset on multi-class log loss in an imbalanced population

I have sampled data and labeled it as being 1 of 14 classes. This dataset is very imbalanced, e.g. I have a lot of samples for class 1 and not that many for class 14. However, this same imbalance is ...
0
votes
1answer
40 views

Unbalanced dataset classification problem

I have a binary classification problem and I'm working with an unbalanced dataset. The count for each class in the training set looks like: ...
0
votes
0answers
14 views

How to tune an weighted voting ensemble method?

I am working on kidney cancer patients' data with 5 unbalanced labels. These codes are contained of Normalization, Oversampling on Feature Engineering part. A list of 9 ordinary Machine Learning ...
0
votes
0answers
17 views

I wonder why the precision of 1 class is much lower on the test set relative to the training set?

I used the LightGBM algorithm to predict the outcome of tennis matches. Next I made two confusion matrices, one for the training set and one for the test set. I calculated stats as precision, recall, ...
1
vote
0answers
19 views

Will very uneven sample sizes within a factor variable cause problems when running a binomial glm? [duplicate]

I'm running a binomial GLM in R. The data for the model comes from survey responses. The response variable is 'change in wellbeing' and the predictor variables are derived from several other questions ...
0
votes
0answers
12 views

crossvalidation “balancing” for regression problems

Classification problems can exhibit a strong label imbalance in the given dataset. This can be overcome by subsampling certain class weight attributed weights, which allow for balancing the label ...
0
votes
1answer
34 views

Classification of Imbalanced and Streaming Time Series Data

I have a question about classification of time series. Data has two features and I want to classify it into 5 classes. We have a stream of data and new data is generated every 5 seconds. Moreover in ...
0
votes
0answers
18 views

What are the measures to detect underrepresented and over represented classes in an imbalanced dataset

In case of imbalanced datasets, we suppose we have the dataset with the corresponding class frequencies ...
0
votes
0answers
13 views

how can i plot a gini curve?

i am using a scoring metric as below: (gini) ...
2
votes
0answers
20 views

Why does the Youden rule does not recommend a threshold of 0.5 on balanced data?

Suppose I have a logistic regression model estimated using a balanced target (equal group sizes). My questions concern the optimal threshold for prediction and it's relationship with the Youden's rule ...
0
votes
0answers
13 views

how can i handle class imbalance from different data sets?

I have an imbalanced problem and my datasets are from 4 different countries; 1 of these countries has % of cancer patients at 2%, the rest are at 1%. When i ran a training job using xgboost it seemed ...
0
votes
0answers
16 views

overfitting and brier score

I have a imbalanced classification problem where i want to see if a client is defaulter or non defaulter. What is important to me is the probability of default, and how well calibrated the model is so ...
1
vote
1answer
47 views

Can oversampling be moved outside stratified k-fold CV?

In a binary classification task, I am using imbalanced-learn's implementation of SMOTENC to oversample the positive class of a very imbalanced dataset. The total number of examples is very high, so ...
0
votes
1answer
32 views

GLS (nlme) model with interaction and unbalanced data

I have a dataset of soil pH data and two explanatory factors, one with 5 levels (land use)and another one with >100 levels (soil type). Data exploration and comparison of several GLS models fit ...
0
votes
0answers
126 views

Constructing a model with SMOTE and sklearn pipeline

I have a very imbalanced dataset on which I'm trying to construct a LinearSVC model with SMOTE and standardization, using a Pipeline. I had already applied SMOTE and sklearn's StandardScaler with ...
0
votes
0answers
5 views

Can tests for balanced designs be used when groups have very similar (maximum difference of 2) but unequal sample sizes?

Example: 4 groups of following sample sizes: 21,20,20,19 Given that the difference in the sample sizes is relatively small, can tests which are suitable for balanced designs still be applied?
1
vote
2answers
97 views

Machine learning strategy for imbalanced data with high number of examples

I am working on a classification problem, with unbalanced classes : Number of positive examples: ~200k; Number of negative ...
3
votes
1answer
43 views

Ramifications of small + unbalanced group sizes, small number of groups for fixed & random effects models?

I have a variable (call it 'group') that I would like to treat as a random effect in a logistic regression. However, the number of groups is small (9 groups, larger than the recommended absolute ...
1
vote
0answers
22 views

Modeling rare event

I'm working with a dataset with roughly 14K records, and am trying to predict a binary variable (death). While the total dataset size is sufficient, the patients with deaths I'm trying to predict make ...
0
votes
0answers
9 views

How to deal with different proportions of positive samples in train vs. test data for logistic regression?

I want to train a logistic regression model with random effects. In training data there is only 0.5% of positive individuals, whereas in test set there is around 5% of positives. I want to train a ...
0
votes
0answers
25 views

Imbalanced Dataset leads to worse results after using balancing methods

I have a very imbalanced Dataset. It's a binary classification. In the train set I have 150,000 times class 0 and 500 times class 1. That's about 0.33% When I train a model like DecisionTree I get a ...
0
votes
1answer
24 views

Random forest hyperparameter to control misclassification

May I know what hyperparameter to tune for random forest classifier to control misclassification? I'm doing a 5-class classification problem and it turns out that most classes are been misclassified ...
3
votes
2answers
39 views

Should we use AUC as an indicator of overfitting when dataset is highly imbalanced?

In my problem, there are 2 class labels, but one label only counts for 1% of the total data. I first divided my data set by train_test_split such that only 10% are test set, then I performed 10-Fold ...
1
vote
0answers
21 views

Unequal sample sizes

I have 11 years of data ranging from May-November. Based off of environmental data, I broke these months into two "seasons". I called these seasons summer (May-September) and fall (October ...
0
votes
0answers
21 views

What role does the true distribution of data play in machine learning?

It seems to be a very simple question: Does the "true distribution" or "natural distribution" of training data matter in machine learning? The motivation for asking this question ...
0
votes
0answers
14 views

How can I improve inspection of the performance of my deep image binary classifier when I have additional data for my classes?

I have some image and also some other attributes (metadata) for each image describing the conditions. My trained classifier's input is only images, and I want to debug the weak points of it using the ...

1
2 3 4 5
17