scholarly journals An insight into the effects of class imbalance and sampling on classification accuracy in credit risk assessment

2019 ◽  
Vol 16 (1) ◽  
pp. 155-178 ◽  
Author(s):  
Kristina Andric ◽  
Damir Kalpic ◽  
Zoran Bohacek
Keyword(s):  
Credit Risk ◽  
Sample Size ◽  
Real Life ◽  
Class Imbalance ◽  
Performance Measure ◽  
Classification Algorithms ◽  
Data Sets ◽  
Valuable Insight ◽  
Data Set ◽  
Class Distribution

In this paper we investigate the role of sample size and class distribution in credit risk assessments, focusing on real life imbalanced data sets. Choosing the optimal sample is of utmost importance for the quality of predictive models and has become an increasingly important topic with the recent advances in automating lending decision processes and the ever growing richness in data collected by financial institutions. To address the observed research gap, a large-scale experimental evaluation of real-life data sets of different characteristics was performed, using several classification algorithms and performance measures. Results indicate that various factors play a role in determining the optimal class distribution, namely the performance measure, classification algorithm and data set characteristics. The study also provides valuable insight on how to design the training sample to maximize prediction performance and the suitability of using different classification algorithms by assessing their sensitivity to class imbalance and sample size.

PSVI-8 Meta-regression Analysis to Determine the Relationship Between Growing Pig Body Weight and Variation

2021 ◽  
Vol 99 (Supplement_1) ◽  
pp. 218-219
Author(s):  
Andres Fernando T Russi ◽  
Mike D Tokach ◽  
Jason C Woodworth ◽  
Joel M DeRouchey ◽  
Robert D Goodband ◽  
...  
Keyword(s):  
Body Weight ◽  
Regression Analysis ◽  
Sample Size ◽  
Polynomial Regression ◽  
Data Sets ◽  
Regression Equations ◽  
Prediction Equations ◽  
Data Set ◽  
Rate Of Increase ◽  
The Relationship

Abstract The swine industry has been constantly evolving to select animals with improved performance traits and to minimize variation in body weight (BW) in order to meet packer specifications. Therefore, understanding variation presents an opportunity for producers to find strategies that could help reduce, manage, or deal with variation of pigs in a barn. A systematic review and meta-analysis was conducted by collecting data from multiple studies and available data sets in order to develop prediction equations for coefficient of variation (CV) and standard deviation (SD) as a function of BW. Information regarding BW variation from 16 papers was recorded to provide approximately 204 data points. Together, these data included 117,268 individually weighed pigs with a sample size that ranged from 104 to 4,108 pigs. A random-effects model with study used as a random effect was developed. Observations were weighted using sample size as an estimate for precision on the analysis, where larger data sets accounted for increased accuracy in the model. Regression equations were developed using the nlme package of R to determine the relationship between BW and its variation. Polynomial regression analysis was conducted separately for each variation measurement. When CV was reported in the data set, SD was calculated and vice versa. The resulting prediction equations were: CV (%) = 20.04 – 0.135 × (BW) + 0.00043 × (BW)2, R2=0.79; SD = 0.41 + 0.150 × (BW) - 0.00041 × (BW)2, R2 = 0.95. These equations suggest that there is evidence for a decreasing quadratic relationship between mean CV of a population and BW of pigs whereby the rate of decrease is smaller as mean pig BW increases from birth to market. Conversely, the rate of increase of SD of a population of pigs is smaller as mean pig BW increases from birth to market.

scholarly journals Examination of Different Item Response Theory Models on Tests Composed of Testlets

2017 ◽  
Vol 6 (4) ◽  
pp. 113
Author(s):  
Esin Yilmaz Kogar ◽  
Hülya Kelecioglu
Keyword(s):  
Item Response Theory ◽  
Sample Size ◽  
Item Response ◽  
Data Sets ◽  
Meaningful Difference ◽  
Response Theory ◽  
Size Change ◽  
Data Set ◽  
Testlet Response Theory ◽  
Item Response Theory Models

The purpose of this research is to first estimate the item and ability parameters and the standard error values related to those parameters obtained from Unidimensional Item Response Theory (UIRT), bifactor (BIF) and Testlet Response Theory models (TRT) in the tests including testlets, when the number of testlets, number of independent items, and sample size change, and then to compare the obtained results. Mathematic test in PISA 2012 was employed as the data collection tool, and 36 items were used to constitute six different data sets containing different numbers of testlets and independent items. Subsequently, from these constituted data sets, three different sample sizes of 250, 500 and 1000 persons were selected randomly. When the findings of the research were examined, it was determined that, generally the lowest mean error values were those obtained from UIRT, and TRT yielded a mean of error estimation lower than that of BIF. It was found that, under all conditions, models which take into consideration the local dependency have provided a better model-data compatibility than UIRT, generally there is no meaningful difference between BIF and TRT, and both models can be used for those data sets. It can be said that when there is a meaningful difference between those two models, generally BIF yields a better result. In addition, it has been determined that, in each sample size and data set, item and ability parameters and correlations of errors of the parameters are generally high.

scholarly journals A new concordant partial AUC and partial c statistic for imbalanced data in the evaluation of machine learning algorithms

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
André M. Carrington ◽  
Paul W. Fieguth ◽  
Hammad Qazi ◽  
Andreas Holzinger ◽  
Helen H. Chen ◽  
...  
Keyword(s):  
Roc Curve ◽  
Diagnostic Testing ◽  
Real Life ◽  
Imbalanced Data ◽  
Machine Learning Algorithms ◽  
Data Sets ◽  
Data Set ◽  
Partial Auc ◽  
C Statistic ◽  
Future Work

Abstract Background In classification and diagnostic testing, the receiver-operator characteristic (ROC) plot and the area under the ROC curve (AUC) describe how an adjustable threshold causes changes in two types of error: false positives and false negatives. Only part of the ROC curve and AUC are informative however when they are used with imbalanced data. Hence, alternatives to the AUC have been proposed, such as the partial AUC and the area under the precision-recall curve. However, these alternatives cannot be as fully interpreted as the AUC, in part because they ignore some information about actual negatives. Methods We derive and propose a new concordant partial AUC and a new partial c statistic for ROC data—as foundational measures and methods to help understand and explain parts of the ROC plot and AUC. Our partial measures are continuous and discrete versions of the same measure, are derived from the AUC and c statistic respectively, are validated as equal to each other, and validated as equal in summation to whole measures where expected. Our partial measures are tested for validity on a classic ROC example from Fawcett, a variation thereof, and two real-life benchmark data sets in breast cancer: the Wisconsin and Ljubljana data sets. Interpretation of an example is then provided. Results Results show the expected equalities between our new partial measures and the existing whole measures. The example interpretation illustrates the need for our newly derived partial measures. Conclusions The concordant partial area under the ROC curve was proposed and unlike previous partial measure alternatives, it maintains the characteristics of the AUC. The first partial c statistic for ROC plots was also proposed as an unbiased interpretation for part of an ROC curve. The expected equalities among and between our newly derived partial measures and their existing full measure counterparts are confirmed. These measures may be used with any data set but this paper focuses on imbalanced data with low prevalence. Future work Future work with our proposed measures may: demonstrate their value for imbalanced data with high prevalence, compare them to other measures not based on areas; and combine them with other ROC measures and techniques.

scholarly journals Precision-Recall versus Accuracy and the Role of Large Data Sets

2019 ◽  
Vol 33 ◽  
pp. 4039-4048 ◽  
Author(s):  
Brendan Juba ◽  
Hai S. Le
Keyword(s):  
Machine Learning ◽  
Class Imbalance ◽  
Imbalanced Data ◽  
Large Data ◽  
Constant Factor ◽  
Data Sets ◽  
Data Set ◽  
Small Constant ◽  
Classifier Performance ◽  
Necessary And Sufficient

Practitioners of data mining and machine learning have long observed that the imbalance of classes in a data set negatively impacts the quality of classifiers trained on that data. Numerous techniques for coping with such imbalances have been proposed, but nearly all lack any theoretical grounding. By contrast, the standard theoretical analysis of machine learning admits no dependence on the imbalance of classes at all. The basic theorems of statistical learning establish the number of examples needed to estimate the accuracy of a classifier as a function of its complexity (VC-dimension) and the confidence desired; the class imbalance does not enter these formulas anywhere. In this work, we consider the measures of classifier performance in terms of precision and recall, a measure that is widely suggested as more appropriate to the classification of imbalanced data. We observe that whenever the precision is moderately large, the worse of the precision and recall is within a small constant factor of the accuracy weighted by the class imbalance. A corollary of this observation is that a larger number of examples is necessary and sufficient to address class imbalance, a finding we also illustrate empirically.

scholarly journals A Novel Approach for Handling Outliers in Imbalanced Data

2018 ◽  
Vol 7 (3.1) ◽  
pp. 1 ◽  
Author(s):  
Gillala Rekha ◽  
V Krishna Reddy
Keyword(s):  
Class Imbalance ◽  
Training Data ◽  
Classification Algorithms ◽  
Main Concern ◽  
Data Generation ◽  
Class Imbalance Problem ◽  
Class Distribution ◽  
Novel Approach ◽  
Real World Datasets ◽  
Data Level

Most of the traditional classification algorithms assume their training data to be well-balanced in terms of class distribution. Real-world datasets, however, are imbalanced in nature thus degrade the performance of the traditional classifiers. To solve this problem, many strategies are adopted to balance the class distribution at the data level. The data level methods balance the imbalance distribution between majority and minority classes using either oversampling or under sampling techniques. The main concern of this paper is to remove the outliers that may generate while using oversampling techniques. In this study, we proposed a novel approach for solving the class imbalance problem at data level by using modified SMOTE to remove the outliers that may exist after synthetic data generation using SMOTE oversampling technique. We extensively compare our approach with SMOTE, SMOTE+ENN, SMOTE+Tomek-Link using 9 datasets from keel repository using classification algorithms. The result reveals that our approach improves the prediction performance for most of the classification algorithms and achieves better performance compared to the existing approaches.   

scholarly journals Feature Scaling via Second-Order Cone Programming

2016 ◽  
Vol 2016 ◽  
pp. 1-7
Author(s):  
Zhizheng Liang
Keyword(s):  
Simulated Data ◽  
Performance Measure ◽  
Second Order ◽  
Data Sets ◽  
Scaling Factors ◽  
Data Set ◽  
Cone Programming ◽  
Second Order Cone Programming ◽  
Second Order Cone ◽  
Feature Scaling

Feature scaling has attracted considerable attention during the past several decades because of its important role in feature selection. In this paper, a novel algorithm for learning scaling factors of features is proposed. It first assigns a nonnegative scaling factor to each feature of data and then adopts a generalized performance measure to learn the optimal scaling factors. It is of interest to note that the proposed model can be transformed into a convex optimization problem: second-order cone programming (SOCP). Thus the scaling factors of features in our method are globally optimal in some sense. Several experiments on simulated data, UCI data sets, and the gene data set are conducted to demonstrate that the proposed method is more effective than previous methods.

A Dynamic Genetic Algorithm for Clustering Problems

2013 ◽  
Vol 411-414 ◽  
pp. 1884-1893
Author(s):  
Yong Chun Cao ◽  
Ya Bin Shao ◽  
Shuang Liang Tian ◽  
Zheng Qi Cai
Keyword(s):  
Genetic Algorithm ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Real Life ◽  
Search Space ◽  
Adaptive Mutation ◽  
Data Sets ◽  
Data Set ◽  
Local Optima ◽  
Clustering Problems

Due to many of the clustering algorithms based on GAs suffer from degeneracy and are easy to fall in local optima, a novel dynamic genetic algorithm for clustering problems (DGA) is proposed. The algorithm adopted the variable length coding to represent individuals and processed the parallel crossover operation in the subpopulation with individuals of the same length, which allows the DGA algorithm clustering to explore the search space more effectively and can automatically obtain the proper number of clusters and the proper partition from a given data set; the algorithm used the dynamic crossover probability and adaptive mutation probability, which prevented the dynamic clustering algorithm from getting stuck at a local optimal solution. The clustering results in the experiments on three artificial data sets and two real-life data sets show that the DGA algorithm derives better performance and higher accuracy on clustering problems.

scholarly journals Exploration of Obesity Status of Indonesia Basic Health Research 2013 With Synthetic Minority Over-Sampling Techniques

2021 ◽  
Vol 5 (1) ◽  
pp. 75-91
Author(s):  
Sri Astuti Thamrin ◽  
Dian Sidik ◽  
Hedi Kuswanto ◽  
Armin Lawi ◽  
Ansariadi Ansariadi
Keyword(s):  
Machine Learning ◽  
Health Research ◽  
Class Imbalance ◽  
Imbalanced Data ◽  
Sampling Technique ◽  
Data Sets ◽  
Data Set ◽  
Basic Health ◽  
Obesity Status ◽  
Obese Class

The accuracy of the data class is very important in classification with a machine learning approach. The more accurate the existing data sets and classes, the better the output generated by machine learning. In fact, classification can experience imbalance class data in which each class does not have the same portion of the data set it has. The existence of data imbalance will affect the classification accuracy. One of the easiest ways to correct imbalanced data classes is to balance it. This study aims to explore the problem of data class imbalance in the medium case dataset and to address the imbalance of data classes as well. The Synthetic Minority Over-Sampling Technique (SMOTE) method is used to overcome the problem of class imbalance in obesity status in Indonesia 2013 Basic Health Research (RISKESDAS). The results show that the number of obese class (13.9%) and non-obese class (84.6%). This means that there is an imbalance in the data class with moderate criteria. Moreover, SMOTE with over-sampling 600% can improve the level of minor classes (obesity). As consequence, the classes of obesity status balanced. Therefore, SMOTE technique was better compared to without SMOTE in exploring the obesity status of Indonesia RISKESDAS 2013.

scholarly journals Performance Test on Classification Algorithms

2019 ◽  
Vol 8 (2S3) ◽  
pp. 914-917
Keyword(s):  
Performance Measures ◽  
Performance Test ◽  
Confusion Matrix ◽  
Weather Data ◽  
Classification Algorithms ◽  
Data Sets ◽  
Huge Amount ◽  
Data Set ◽  
Random Forest Classification ◽  
Forest Classification

Nowadays, a huge amount of data is generated due to the growth in the technologies. There are different tools used to view this massive amount of data, and these tools contain different data mining techniques which can be applied for the obtained data sets. Classification is required to extract useful information or to predict the result from these enormous amounts of data. For this purpose, there are different classification algorithms. In this paper, we have compared Naive Bayes, K*, and random forest classification algorithm using Weka tool. To analyze the performance of these three algorithms we have considered three data sets. They are diabetes, supermarket and weather data set. In this work, an analysis is made based on the confusion matrix and different performance measures like RMSE, MAE, ROC, etc

Sign in / Sign up

Export Citation Format

Share Document