the loss(passive) caused bythe loss

the plausible bias(passive) caused bythe plausible bias

to seemingly good results despite a lack of separability ... for instance the differentiation of luminal B vs. all others ( 75 %ledto seemingly good results despite a lack of separability ... for instance the differentiation of luminal B vs. all others ( 75 %

fewer problems for regression techniques than for classifiers because in the regression model , the intercept value moves the outcome of the regression function towards the bias in the datacausesfewer problems for regression techniques than for classifiers because in the regression model , the intercept value moves the outcome of the regression function towards the bias in the data

to a learning bias towards the majority classcan leadto a learning bias towards the majority class

a bias towards the majority class ... 64can causea bias towards the majority class ... 64

poor performance as algorithms tend to classify majority class examples better in expense of minority class cases as the total misclassification error is much improved when majority class is labeled correctlycausespoor performance as algorithms tend to classify majority class examples better in expense of minority class cases as the total misclassification error is much improved when majority class is labeled correctly

on the covering operatorcauseon the covering operator

a learning algorithm during training by making decision rule biased towards the majority class and optimizes the predictions based on the majority class in the datasetinfluencesa learning algorithm during training by making decision rule biased towards the majority class and optimizes the predictions based on the majority class in the dataset

a learning algorithm during training by making decision rules biased towards the majority class and optimizing the predictions based on the majority class in the datasetinfluencesa learning algorithm during training by making decision rules biased towards the majority class and optimizing the predictions based on the majority class in the dataset

deep problemscan createdeep problems

from an unequal amount of tumor and normal tissue in the dataresultingfrom an unequal amount of tumor and normal tissue in the data

ROC curves to be poor visualizations of classifier performancecan causeROC curves to be poor visualizations of classifier performance

features(passive) caused byfeatures

to either more class - to - class animosity or the need for further homogenizationleadsto either more class - to - class animosity or the need for further homogenization

churn models to break down because of the lack of informationmay causechurn models to break down because of the lack of information

rules of the minority class When XCS will remove these rules Population size bound with respect to the imbalance ratio Until which imbalance ratio would XCS be able to learn from the minority classcan ... createrules of the minority class When XCS will remove these rules Population size bound with respect to the imbalance ratio Until which imbalance ratio would XCS be able to learn from the minority class

two problems : ( 1 ) training is inefficient as most locations are easy negatives that contribute no useful learning signal ; ( 2 ) the easy negatives can overwhelm training and lead to degenerate modelscausestwo problems : ( 1 ) training is inefficient as most locations are easy negatives that contribute no useful learning signal ; ( 2 ) the easy negatives can overwhelm training and lead to degenerate models

to some troubles for data mining algorithms assuming an almost equal class distributionleadsto some troubles for data mining algorithms assuming an almost equal class distribution

to these high numbers ... though AUC is more robust to class imbalance than other measures ( i.e. , accuracymay have contributedto these high numbers ... though AUC is more robust to class imbalance than other measures ( i.e. , accuracy

PR - AUC(passive) is ... influenced byPR - AUC

the discarding of the negative samples ... leaving only positive class samples in the training data setcould causethe discarding of the negative samples ... leaving only positive class samples in the training data set

the small disjunct problem(passive) created bythe small disjunct problem

to low test scorescontributesto low test scores

the choice of a suitable performance measureinfluencesthe choice of a suitable performance measure

a learning algorithm during training by making the decision rule biased towards the majority class by implicitly learns a model that optimizes the predictions based on the majority class in the datasetinfluencesa learning algorithm during training by making the decision rule biased towards the majority class by implicitly learns a model that optimizes the predictions based on the majority class in the dataset

to the cold - start problem , which occurs in situations where decisions or historical data are requiredcontributesto the cold - start problem , which occurs in situations where decisions or historical data are required

in higher uncertainty for one groupresultingin higher uncertainty for one group

some issues with choosing the right metric to evaluate ( so much so that the evaluation metric for this competition was actually changed mid - competition from cross - entropy to AUCcan causesome issues with choosing the right metric to evaluate ( so much so that the evaluation metric for this competition was actually changed mid - competition from cross - entropy to AUC

a narrower pvp class spreadwould have createda narrower pvp class spread

into the followingresultsinto the following

AUC variability(passive) caused byAUC variability

the difficulty of predictioncausingthe difficulty of prediction

the effect(passive) caused bythe effect

a neural network ’s stochastic gradient descent traininginfluencesa neural network ’s stochastic gradient descent training

along with the challenging nature of the datasetleadalong with the challenging nature of the dataset

the resulting probabilities from logistic regression modelsinfluencesthe resulting probabilities from logistic regression models

to the improved recall scores for the positive classleadsto the improved recall scores for the positive class

the program 's interest in junior college transfershas sparkedthe program 's interest in junior college transfers