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
1answer
29 views

Removing duplicate training vectors?

As an extension to this question, for ML problems where it makes sense to remove duplicates (ie: identical data & target variables) from your distribution, in which scenarios would it (if at all) ...
1
vote
1answer
21 views

F1 Score is giving good value in imbalanced dataset

If I have an imbalanced dataset that consists of 90% positive points and 10% negative points. Now I created a "dumb" model which always predicts every point as a positive point. The ...
2
votes
1answer
30 views

Outliers for classification; imbalanced data for regression?

When I hear of the terms "outliers" and "imbalanced data" it's usually in the context of regression and classification respectively. With "outliers" meaning the ...
1
vote
0answers
19 views

When we up-sample the training set, don't we introduce selection bias?

When doing supervised machine learning in the health or medical domains, we often have a target class that is relatively rare (e.g., prevalence 1-10% of cases). There are a few techniques we can do to ...
0
votes
0answers
16 views

Forcing uniform prior when training classifier?

Say you're training a classifier to take an input $x$ and predict its label $y \in \{1,\ldots,k\}$. As an example, let's say the classifier is a neural net, which ends in a softmax layer, and we train ...
1
vote
0answers
18 views

Multivariant classification imbalance dataset [closed]

The dataset head looks like this. ( working in RStudio) head(df) ...
0
votes
1answer
24 views

How to increase estimated probability of minority class in imbalance data?

If I use model that provide output probability in imbalanced case (say ratio between majority and minority class is 100 : 1), I saw that the output probability of data points from majority class is ...
3
votes
0answers
175 views

Resampling classes across weighted source distributions

I am sure this is a common problem, but googling only yielded false positives. I probably did not know what terms to search for. So here we go: I have $n$ classes from $m$ different sources. Each ...
0
votes
0answers
7 views

Define sampling rate in the context of SMOTE

In this paper, authors claimed that the traditional SMOTE uses the same sampling rate for all instances of the minority class whereas their proposed genetic algorithm-based SMOTE (GASMOTE) algorithm ...
10
votes
3answers
228 views

How do you know that your classifier is suffering from class imbalance?

In cases where there is a substantial difference in relative class frequencies, it could be that the density of the minority class is never higher than the density of the majority class anywhere in ...
1
vote
1answer
47 views

What can i use to fix SVM overfitting when all expected solutions have failed?

I am trying to predict if a certain day will be good for agriculture based off a select number of features, the data has an imbalance of 5:1. There's a total of 1794 samples and 15 variables, columns ...
0
votes
0answers
31 views

Mixture Density Network on Unbalanced Dataset

I want to use a mixture density network to make predictions with NLL loss (which is working) but my dataset is unbalanced so my network tends to make estimations that look like the distribution of ...
1
vote
1answer
47 views

Measuring Precision/Recall on a biased sample

I am working with ML models that predict e.g. whether an email violates some corporate policy or not. In this case, the "positives" are emails that violate the policy, and the number of ...
0
votes
0answers
67 views

Analysis of variance with unbalanced design and heteroskedastic data

I am looking for a way to analyse the differences in vegetation carbon stocks (named after 'OC_TOT') among different vegetation types (named after 'Facies') and islands (Guadeloupe and Martinique). ...
1
vote
0answers
26 views

Weighted vs Non weighted scores for unbalanced classes

I have a dataset which has 99.8% negatives and 0.2% positives. The scores I have got for my model(built using XGBoost) is as follows: ...
1
vote
0answers
50 views

R: Handling unbalanced data for Random Forest regression

I have a regression problem that I solve with a random forest in R. Input are several predictors (up to 40), output is a point in time (represented as a year between 1867 and 2017). The reference data ...
0
votes
0answers
14 views

How to handle class imbalance in classification with same feature values

I am trying to build a model to predict click through rate for advertisement. I have the data in the following format ...
1
vote
1answer
41 views

upsampling vs class weights in mini-batch SGD

Let's consider using mini-batch SGD in (neural network) binary classification problem with imbalanced dataset. Let's say that the ratio between the number of examples in each class is positive:...
0
votes
2answers
39 views

How to choose between 2 strategies to train a Deep Learning model on an unbalanced Dataset?

I have a Training Set of respiratory disease sounds, so there are 2 classes: 0 for respiratory sounds of healthy patients. 1 for breathing sounds of patients with a disease. The Training Set is ...
1
vote
1answer
108 views

Imbalanced data to match reality with Random Forest?

I have a medical dataset which I am using select features to predict for 1 of 3 diseases. I have found that Random Forest works best for my dataset after testing out various combinations. The original ...
0
votes
0answers
34 views

Overfitting in Gradient Boosting

I have a very unbalanced dataset(99.8% negative,0.2% positive) with approximately 60 variables. I removed somewhere around 40 variables based on the variance inflation factor. Then I used SMOTE to ...
1
vote
0answers
20 views

Help with understanding metrics for imbalanced classification

I am trying to train a neural network to classify chest X-ray scans as my final MSc project. I have a dataset of 13808 image, 3616 labelled COVID, 10192 labelled normal, so the ratio of COVID to ...
0
votes
1answer
25 views

loss function for probability maps [duplicate]

I am applying a deep learning architecture to predict the spatial distribution of nightly thunderstorms over location B based on the thunderstorms of the preceding afternoon over location A. Their ...
0
votes
0answers
18 views

In classification algorithms, does the probabilty changes depending on how you define the label? and if is unbalanced?

I have been struggling with this topic for a while, and on this site are multiple answers but none of them answers completely my question. For example in Which class to define as positive in ...
11
votes
2answers
661 views

When should I balance my data for GLM/GAM?

I'm running GLMs and GAMs and I can't find a clear answer about if I should be balancing my data or not. I'm trying to get useful descriptive models, not predictive models. So I haven't split my data ...
5
votes
2answers
238 views

Which class to define as positive in unbalanced classification

In a classification task, why do we usually choose the minority class as "positive" case for response variable? For example, if there are 1000, 9000 for class A and B respectively, we ...
0
votes
1answer
18 views

Selecting a label smoothing factor for seq2seq NMT with a massive imbalanced vocabulary

I'm training a seq2seq RNN with a vocabulary of 8192 words. This means that the typical categorical cross entropy label smoothing factor suggested in papers like 'Attention is all you need' of $0.1$ ...
0
votes
1answer
19 views

How do unbalanced classes in binary predictors affect the regression of a continous or binary outcome?

I would like to learn more about how unbalanced binary predictors affect type I or type II error in regression models. I am aware that having unbalanced binary outcomes can increase type II error. I ...
1
vote
0answers
30 views

Evaluation metric for imbalanced data

Hi I'm a CS graduate student I have a question for AI or data experts. I'm writing a paper My dataset is time-series sensor data and anomaly (positive class) ratio is between 5% and 6% you can see the ...
0
votes
0answers
12 views

Is there any way to oversample mixed data?

I am working in a binary classification problem and I have two inputs to the network (a dataset and images). In the first branch I use a Multi-layer Perceptron (MLP) to handle the dataset and in the ...
0
votes
0answers
15 views

Evaluating a CNN -multi class model with two separate thresholds

I have a model that outputs three classes. But here instead of one threshold, it depends on a combination of two (user input threshold). One threshold varies from 0.1 to 1.0 and the other varies from ...
0
votes
0answers
27 views

Comparing ROC AUPRC scores in case of different baselines

I have some imbalanced data for binary classification, which I have preprocessed in 2 different ways. That led to having a different number of observations and pos/neg ratio. Then I trained the same ...
0
votes
0answers
48 views

Binary classification, F1 score 0=0.95, 1=0.06

I'm building a machine learning model to predict a process failure (1=fail, 0=no-fail). To begin with, I have a class imbalance ratio of 1:51. After some wrangling, I applied Clustering-Based ...
0
votes
0answers
7 views

SMOTEBoost implementation

This question got to do with SMOTEBoost implementation found here but I believe the issue is relayed to imblearn library. I ...
0
votes
1answer
26 views

CoxTimeVaryingFitter model for Inference

I am trying to use CoxTimeVaryingFitter model in python from lifelines package, for making inference on which features have a causal impact on a success outcome. The features are time-varying so this ...
0
votes
1answer
50 views

Training a random forest with an imbalanced dataset with ranger, which parameter to use for weights?

I would like to train a random forest and I got an unbalanced dataset which could look like this: class number of observations weight 0 20 1 1 10 2 2 5 4 In order to take the dataset imbalance ...
2
votes
1answer
100 views

Is threshold moving unnecessary in balanced classification problem?

As far as I know, the threshold moving is needed in imbalanced classification problems. The reason why we have to adjust the decision threshold is as follows: Most machine learning algorithms are ...
1
vote
0answers
13 views

Parameter tuning in classification

Recently, I am digging into the selection of tuning parameters in a binary classification. I gathered information by googling and the following is my organization. We can distinct binary ...
0
votes
0answers
14 views

Evaluating a convolutional neural network on an imbalanced (academic) dataset

I have trained a posture analysis network to classify in a video of humans recorded in public places if there is a) shake-hand between two humans, b) Standing close together that their hands touch ...
0
votes
0answers
19 views

How to avoid having only one class during cross-validation

I have to perform a binary classification. My dataset is quite small 280 samples and quite imbalanced (1:10 ratio). I kept around 100 sample as testing and about 140 for training. My input variables ...
2
votes
1answer
46 views

Sample size confounded with factors (ANOVA)

What do you suggest doing when sample size is confounded with factors in an ANOVA? "For example, in a two-way ANOVA, let’s say that your two independent variables (factors) are Age (young vs. old)...
0
votes
0answers
8 views

Finding and optimising a three group classification criteria for a $\{0,1\}$ response with one independent variable

I want to create a grouping algorithm that groups my data into $3$ classes. Now imagine that we have one independent variable $X$, which can be either discrete or real, and a dependent variable $Y\in \...
1
vote
0answers
25 views

validation and test set definition for active learing with rare classes

Context I've got an active learning problem with an event rate of about 1%. The data is a panel, individuals over time. We have a proxy label that is highly correlated with the true label within ...
7
votes
1answer
364 views

Biased estimates in logistic regression due to class imbalance

I was asked by a reviewer to evaluate the robustness of the results of logistic regression, given that estimates can be biased by class imbalance in the outcome. To contextualize, I have run three ...
1
vote
1answer
14 views

How does Matthews Correlation Coefficient define random predictions for binary classificattion?

As stated in Matthews, Brian W. "Comparison of the predicted and observed secondary structure of T4 phage lysozyme." C = 0 is expected for a prediction no better than random I'm not sure ...
0
votes
0answers
89 views

What should be the class imbalance ratio?

I'm working with a really imbalanced binary classification dataset so I decided to use SMOTE for only on the train data. Class rates were 95% -5% before SMOTE and 75-25% after SMOTE. In other words, I ...
0
votes
0answers
30 views

Is balancing class data for imbalanced problems helpful or just folklore when considering thresholds?

(In the context of predictive models) Caveat: I'm aware that imbalanced data questions are a dead horse, but I haven't found an answer to this flavor of it directly. When working with highly ...
0
votes
0answers
8 views

Passing multiple fixed length sequences to LSTM

The task at hand is multi-class text classification. I have 22,806 documents and 103 classes. The documents are of varying length: ...
0
votes
0answers
12 views

Regression model selection and specification for unbalanced repeated measures data

My hypothesis is that person's blood oxygen level is affected by their age, gender, and drugs they have taken in the previous 24 hours. I would like to confirm and quantify effects associated with ...
0
votes
0answers
29 views

Classification evaluation: How to sample the confusion matrix for unbalanced data?

Goal and Background My goal is to classify Twitter accounts: Account of scientist 0 = no, 1 = yes. To this end, a scientist identification algorithm was applied to a total of 7 Mio. accounts (no ...

1
2 3 4 5
18