I've seen this term several times already, and have no idea what it means, can you explain what it means ?
Solved – an imbalanced data set
classificationdatasetunbalanced-classes
Related Solutions
Since you're using R, you could make use of some elaborated methods like ROSE and SMOTE. But I'm not enrirely certain if re-balancing your dataset is the right solution in your case.
An alternative could be a cost-sensitive algorithm like C5.0 that doesn't need balanced data. You could also think about applying Markov chains to your problem.
As your class sizes are so big. I would perform a pre-downsampling to something like 5000+10000+10000+10000+10000. Do you really need more samples? Then downsample again and model independently and aggregate multiple forests afterwards. That will save time and memory. During modeling you may even only bootstrap ~5000 samples for each tree to speedup process. For each tree the bootstrap can be stratified, such that 1000 samples from each class are selected.
Here's a thread on how to train a balanced multi class forest with down sampling and 1-vs-rest ROC plot.
And here's a R-code example on 1-vs-rest roc plots:
library(AUC)
#simulated probabilistic prediction(yhat) vs true class (y)
obs=500
nClass=5
y = sample(1:nClass,obs,rep=T)
yhat = sapply(y,function(y) {
pred.prob = rep(0,nClass)
pred.prob[y] = 0.2
pred.prob = pred.prob + runif(nClass)
pred.prob = pred.prob / sum(pred.prob)
})
#plot 1-vs-all, one curve for each class
for(i in 1:nClass) plot(roc(predictions = yhat[i,],
labels = as.factor(y==i)),
add=i!=1,
col=i)
Best Answer
Answer: A dataset is imbalanced if the distribution of classes is not uniform.
in other (paper reference) words: Imbalance problem occur where one of the two classes having more sample than other classes.
Problem: The problem arising from imbalance is that accuracy can sometimes be high for a classifier although it does perform rather bad.
Example: A simple example could be found in disease diagnostics. Assume you have a disease which only a very few people actually have. Lets assume you have a dataset of 100 people. 90 people are healthy, 10 are sick. Thus, if you construct a classifier which always predicts "healthy", you will achieve a wonderful accuracy of 90%. Yet in this specific case it is a rather bad performance because all sick people are falsely diagnosed as healthy.
What to do: Here sensitivity and specificity come in handy which are implicitly used when computing the AUC (area under the curve) for evaluation of a classifier. Wikipedia provides great further information.