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*.

learn more… | top users | synonyms

1
vote
1answer
100 views

Logistic sample and case numbers

I have some questions about binary logistic regression. For my research, I am planning to use 12 predictors, and my sample consists of 129 cases. However, I know of a 1 to 10 rule. Additionally, my ...
1
vote
2answers
251 views

How to deal with unbalanced data

I'm doing data analysis with a dataset of 11795 data points (with 88 features). 85% (9973 points) of these data points correspond to data points belonging to class 1, 5% (589 points) belong to class 2 ...
0
votes
0answers
24 views

Is it correct to compare two distributions of observed (that are means of observation) and expected values with chi square?

I have 2 classification methods that I want to compare. So I ran them on the same dataset and I obtained 2 different classifications. Then, I want to test if their classification is robust and how. ...
1
vote
0answers
100 views

Oversampling with categorical variables

I would like to perform a combination of oversampling and undersampling in order to balance my dataset with roughly 4000 customers divided into two groups, where one of the groups have a proportion of ...
0
votes
0answers
21 views

the effect of unbalanced classes

I have a data set with label "Y" and "N", which is splitted into training and testing sets. The ratio the labels in training and testing sets are 3:1 and 1:1, respectively. If I build model using the ...
0
votes
1answer
78 views

Improving SVM classification

I have a classification problem (bioinformatics domain) where I have around 333 features. Currently, I am first selecting features (using importance feature of random forest) and then pushing the same ...
0
votes
0answers
27 views

How to account for multiple & varying amounts of observations per factor level & still retain info? As a random effect in GLMM or take the mean?

I would appreciate any help! Specifically I would like to know which option is best. Question Does Var2 influence Var1 in relation to the factor(Ind), and does Var3 & Var4 also have some effect. ...
0
votes
0answers
58 views

How to handle imbalanced data using SMOTE-N (nominal data) to generate multiple synthetic data?

I have a classification problem with two classes working on nominal data. I want to apply SMOTE-N to deal with imbalanced data. However, it is not clear to me how to use SMOTE-N for generating N ...
0
votes
0answers
723 views

One-way ANOVA with unequal sample sizes

I have a function for performing one-way ANOVA in R: ...
1
vote
1answer
97 views

Is f-measure synonymous with accuracy?

I understand that f-measure (based on precision and recall) is an estimate of how accurate a classifier is. Also, f-measure is favored over accuracy when we have an unbalanced dataset. I have a simple ...
1
vote
0answers
71 views

Testing classifier with binomial test when group sizes are unequal

I have data from 50 human subjects, who are divided into groups A and B (30 participants are in group A and 20 participants in group B). I also have a range of measurements from each subject. I have ...
0
votes
0answers
25 views

Comparing networks of groups of unequal size

I want to compare some network properties such as entropy, connectivity, etc., between 2 groups. I have time series fMRI data on patients (n = 80) and controls (n = 30) for certain regions of the ...
2
votes
3answers
586 views

Suggestions for cost-sensitive learning in a highly imbalanced setting

I have a dataset with a few million rows and ~100 columns. I would like to detect about 1% of the examples in the dataset, which belong to a common class. I have a minimum precision constraint, but ...
0
votes
2answers
408 views

Handling unbalanced data using SMOTE - NO BIG DIFFERENCE?

I have a classification problem with 2 classes. I have nearly 5000 samples, each of which is represented as vector with 570 features. The positive class samples are nearly 600. Meaning, I have a 1:8 ...
0
votes
0answers
82 views

Repeated Measures ANOVA or mixed model? And with massive group differences?

I need some guidance on which statistical test to use: I want to know if the nature of frog calls changes depending on the time of day. I observed 4 species of frogs at the same 3 times for several ...
0
votes
2answers
89 views

Determine if difference in class distribution is statistically significant

I have a dataset of some observations with class attribute of values 0 and 1. The dataset is quite unbalanced (class 1 – 15%, class 0 – 85%). Further this dataset consists of 5 years, and the ...
2
votes
1answer
223 views

Which metric should I trust to evaluate my predictive model

I am working on predictive model and when I evaluate it, I find good accuracy_score, precision_score, ...
0
votes
0answers
45 views

Trial with two sites, two years - Different number blocks - Problem? (Q1)

I have got two questions on an agricultural field trial that was conducted at two sites in two conscutive years. Virtually everything was the same in all trials (crop variety, planting density...). ...
3
votes
3answers
773 views

SVM for unbalanced data

I want to attempt to use Support Vector Machines (SVMs) on my dataset. Before I attempt the problem though, I was warned that SVMs dont perform well on extremely unbalanced data. In my case, I can ...
2
votes
2answers
70 views

Collecting training data for document classification with unbalanced classes

I have a document classification problem in which the estimated class proportions in the population are severely unbalanced: the population is ~99% class 0 and ~1% class 1. I am using a logistic ...
0
votes
0answers
73 views

How do I partition variance among nested random effects for non-normal data and an unbalanced design?

I have a dataset of plant drought tolerance values (called TLP_DRY) that I would like to partition variance for among the nested levels Biome/Study site/Species to figure out whether most variation in ...
0
votes
1answer
249 views

Should I specify Prior or Cost matrix with Tree Bagger in Matlab

I'm trying to create Random Forests in Matlab and there are more observations in some classes than there are in others. Do I need to specify this as a cost matrix or as a prior probability or will ...
3
votes
1answer
268 views

Using Wilcoxon-Mann-Whitney test for comparing two population of different sizes

I am using Wilcoxon-Mann-Whitney test for comparing two populations. Unfortunately, sizes of my population are different; one has size 100000 and the other 6000. Can I use this test to compare these ...
2
votes
1answer
303 views

unbalanced samples random Forests

I am trying to predict species presence or absence using randomForest in R (classification). In fact, I am trying to do it for several species, in separate models. For a couple of the species, the ...
0
votes
3answers
835 views

Correlation and group difference with unequal sample sizes

My sample is 60 participants. At first, I did not expect to analyze the correlation, but the result is interesting, so I would like to. The 60 participants are 50 people who have good reading skill ...
3
votes
1answer
114 views

Cross Validation in Unbalanced Datasets

Is there a specific way of sampling which maintains the ratio of samples in an unbiased set? e.g., lets say I want to do k-fold cross-validation on my training set And my training set is very ...
4
votes
2answers
3k views

Calculating statistical significance with unequal sample sizes and unequal variances

I have two samples, one with $n_1 = 41,000$ and the other with $n_2 = 881$; the larger sample has a standard deviation of $13.74$, and the smaller has an $SD=10.75$. The means are different, and when ...
1
vote
0answers
152 views

Questionnaire analysis and grouping responses

I want to know if the creative process engagement (CPE) is related to creativity. I don't know how to analyze the questionnaire. I will be collecting data for CPE with Zhang and Bartal's (2010) ...
2
votes
1answer
368 views

How to analyze unequal samples in linear and multiple regression

I have three variables (sample size is mentioned below), and I want to analyze data using regression models as recommended by Judd and Kenny (1981) to see if $cc$ mediates the relationship between ...
3
votes
0answers
209 views

Mixed ANOVA: small and unbalanced samples

I have to analyze two samples ($n_1=10, n_2=18$) in a design in which there is a between-subject factor (Groups: 2 levels) and a within-subjects factor (...
8
votes
2answers
1k views

Does a big difference in sample sizes matter for an independent t-test?

There is a very confusing question in my mind. I have data, and would like to compare numeric scores between men and women. There is a big difference in those two groups: the number of men is 34, ...
0
votes
0answers
140 views

Interpreting coefficients for random effects models with extremely unbalanced data

I'm currently working with a data set that has numerous samples collected over time at different sites in a study area, and I'm interested in detecting a trend over time for that area. I know that in ...
3
votes
1answer
59 views

What test of significance should I use to analyse visits to a clinic by various groups?

Apologies for the noob question. After a day on Google and Wikipedia I still can't quite work out what to do. So here I am. A small private healthcare clinic in the UK has asked me to look at their ...
0
votes
2answers
91 views

Choosing the best featureset for prediction

I have this N sets of features F each with $F_i$ number of features. All the feature sets have 20000 examples and we have 20,000 labels for them. Lets say feature set $F_1$ has 10 features and ...
1
vote
1answer
99 views

Finding the best dataset for classification

I have 100 datasets. All of them have varying number of features. There are around 20,000 samples in each of them. Every $i$-th sample in the 100 datasets has the same label ($0/1$). The data is ...
0
votes
2answers
480 views

How similar are my 2 data sets?

I am kind of stuck with an easy question: I have two data sets with experimental data. The data sets do not have the same size. I would like to show that these data sets are possibly coming from the ...
0
votes
0answers
29 views

Impact of biased sampling on classifier training ?

Let's imagine I have a very unbalanced dataset with 99.99% of 0 and 0.01% of 1 on the target variable. What I want to do is make a classifier for this target. Now imagine that this dataset is very ...
5
votes
2answers
573 views

Classification problem using imbalanced dataset

I am working on a pattern identification/classification problem on an imbalanced dataset, with target to non target proportion in population approx as 1%:99%. There are around 0.5 million records in ...
3
votes
1answer
504 views

classification threshold in RandomForest-sklearn

1) How can I change classification threshold (i think it is 0.5 by default) in RandomForest in sklearn? 2) how can I under-sample in sklearn? 3) I have the following result from RandomForest ...
0
votes
2answers
1k views

Dataset of repeated measures with unbalanced sample size and unequal variace - What to do?

I have a dataset of three groups of cells treated with 10 different compounds and am not sure how to check for significant differences between those treatments. Within each group the data is also ...
0
votes
1answer
365 views

Imbalanced data classification using boosting algorithms

I am working on a binary data classification problem. The dataset is imbalanced, it consists of 92% 'false' labels and 8% 'true' labels. The number of features is 18 and I have a small number of 650 ...
1
vote
1answer
194 views

Denominator is Zero for Matthews correlation coefficient and F-measure

Recently, I built a classification model based on the imbalanced data set(positive sample is minority and negative sample is majority), and the model gave the following result for the test set: ...
4
votes
1answer
609 views

Is a large control sample better than a balanced sample size when the treatment group is small?

I am running an experiment looking at brain volume changes in a rare disorder. We have a small number of patients (n = 8) but a large control group (n = 100). Some colleagues have suggested that a ...
2
votes
1answer
134 views

Outlier detection: at which degree of class imbalance would you consider a one-class model over a two-class model

Background: I am working on the problem of classifying objects found in some biological images. Time and again, we encounter objects which do not fall into any of the categories/classes we are ...
2
votes
0answers
302 views

Repeated measures ANOVA for an experiment with missing values

I have an experiment where several subjects (subjects $= S_1,S_2,...,S_m $) were asked to perform a set of tasks (tasks $= T_1, T_2, T_3,...,T_n$) using both their left ($L$) and right ($R$) arms. ...
2
votes
1answer
488 views

SMOTE throws error for multi class imbalance problem

I am trying to use SMOTE to correct imbalance in my multi-class classification problem. Although SMOTE works perfectly on the iris dataset as per the SMOTE help document, it does not work on a similar ...
3
votes
2answers
309 views

Named entity recognition and class imbalance

I have implemented Maximum-entropy Markov model (MEMM) for the Named entity recognition (NER) problem. I have four classes: geographical, people, material (book titles etc) and other. Class ...
2
votes
2answers
944 views

By using SMOTE the classification of the validation set is bad

I want to do classification with 2 classes. When I classify without smote I get: ...
2
votes
1answer
73 views

Looking for simple examples of how to calculate type III and type IV SS

I have data collected on five species of fish at half a dozen locations in a lake over four years. The categories are not (at all) fully crossed, and I have a lot of empty cells due to logistical ...
4
votes
2answers
558 views

Testing Classification on Oversampled Imbalance Data

I am working on severely imbalanced data. In literature, several methods are used to re-balance the data using re-sampling (over- or under-sampling). Two good approaches are: SMOTE: Synthetic ...