• Journal of Intelligence and Information Systems
• /
• v.15 no.3
• /
• pp.1-15
• /
• 2009

In a classification problem, data imbalance occurs when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. This paper proposes a Geometric Mean-based Boosting (GM-Boost) to resolve the problem of data imbalance. Since GM-Boost introduces the notion of geometric mean, it can perform learning process considering both majority and minority sides, and reinforce the learning on misclassified data. An empirical study with bankruptcy prediction on Korea companies shows that GM-Boost has the higher classification accuracy than previous methods including Under-sampling, Over-Sampling, and AdaBoost, used in imbalanced data and robust learning performance regardless of the degree of data imbalance.

• Korean Journal of Occupational Health Nursing
• /
• v.31 no.2
• /
• pp.77-85
• /
• 2022

Purpose: This study aimed to identify the influence of effort-reward imbalance and job satisfaction on turnover intention among hospital nurses. Methods: Data were collected from January 28 to February 10, 2022, from 237 nurses from five hospitals including clinics, general hospitals, and tertiary care hospitals located in B city. The collected data were analyzed using descriptive statistics, t-test, ANOVA, the Scheffe test, Pearson's correlation coefficients, and multiple linear regression analysis, using SPSS/WIN 26.0. Results: The average of the effort-reward ratio, an indicator of effort-reward imbalance, was 1.67±0.66, and 86.5% of the participants had a value of 1 or more. The mean job satisfaction and turnover intention were 3.32±0.48 and 3.69±0.89 on a 5-point scale, respectively. Multiple regression revealed that factors affecting turnover intention among hospital nurses included effort-reward imbalance (β=.30, p<.001) and job satisfaction (β=-.32, p<.001), and these variables explained 29.0% of turnover intention. Conclusion: These findings indicate that effort-reward imbalance and job satisfaction are associated with turnover intention. Therefore, to decrease the turnover intention of hospital nurses, interventions and policies should be prepared to resolve the nurse's effort-reward imbalance and increase job satisfaction at regional or national level hospitals.

• Information Systems Review
• /
• v.24 no.1
• /
• pp.39-57
• /
• 2022

This paper proposes the GMOPTBoost algorithm to improve the performance of the AdaBoost algorithm for bankruptcy prediction in which class imbalance problem is inherent. AdaBoost algorithm has the advantage of providing a robust learning opportunity for misclassified samples. However, there is a limitation in addressing class imbalance problem because the concept of arithmetic mean accuracy is embedded in AdaBoost algorithm. GMOPTBoost can optimize the geometric mean accuracy and effectively solve the category imbalance problem by applying Gaussian gradient descent. The samples are constructed according to the following two phases. First, five class imbalance datasets are constructed to verify the effect of the class imbalance problem on the performance of the prediction model and the performance improvement effect of GMOPTBoost. Second, class balanced data are constituted through data sampling techniques to verify the performance improvement effect of GMOPTBoost. The main results of 30 times of cross-validation analyzes are as follows. First, the class imbalance problem degrades the performance of ensembles. Second, GMOPTBoost contributes to performance improvements of AdaBoost ensembles trained on imbalanced datasets. Third, Data sampling techniques have a positive impact on performance improvement. Finally, GMOPTBoost contributes to significant performance improvement of AdaBoost ensembles trained on balanced datasets.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.1
• /
• pp.225-243
• /
• 2016

Thanks to recent advance of next generation sequencing techniques, RNA-seq enabled to have an unprecedented opportunity to identify transcript variants with isoform diversity and allelic imbalance (Anders et al., 2012) by different transcriptional rates. To date, it is well known that those features might be associated with the aberrant patterns of disease complexity such as tissue (Anders and Huber, 2010; Anders et al., 2012; Nariai et al., 2014) specific differential expression at isoform levels or tissue specific allelic imbalance in mal-functionality of disease processes, etc. Nevertheless, the knowledge of post-transcriptional modification and AI in transcriptomic and genomic areas has been little known in the traditional platforms due to the limitation of technology and insufficient resolution. We here stress the potential of isoform variability and allelic specific expression that are relevant to the abnormality of disease mechanisms in transcriptional genetic regulatory networks. In addition, we systematically review how robust Bayesian approaches in RNA-seq have been developed and utilized in this regard in the field.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.23-43
• /
• 2022

In order to apply AI (Artificial Intelligence) in various industries, interest in algorithm selection is increasing. Algorithm selection is largely determined by the experience of a data scientist. However, in the case of an inexperienced data scientist, an algorithm is selected through meta-learning based on dataset characteristics. However, since the selection process is a black box, it was not possible to know on what basis the existing algorithm recommendation was derived. Accordingly, this study uses k-means cluster analysis to classify types according to data set characteristics, and to explore suitable classification algorithms and methods for resolving class imbalance. As a result of this study, four types were derived, and an appropriate class imbalance resolution method and classification algorithm were recommended according to the data set type.

• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.75-81
• /
• 2019

Data imbalance problem is common and causes serious problem in machine learning process. Sampling is one of the effective methods for solving data imbalance problem. Over-sampling increases the number of instances, so when over-sampling is applied in imbalanced data, it is applied to minority instances. Under-sampling reduces instances, which usually is performed on majority data. We apply under-sampling and over-sampling to imbalanced data and generate sampled data sets. From the generated data sets from sampling and original data set, we construct a heterogeneous ensemble of classifiers. We apply five different algorithms to the heterogeneous ensemble. Experimental results on an intrusion detection dataset as an imbalanced datasets show that our approach shows effective results.

• Journal of Biomedical Engineering Research
• /
• v.35 no.1
• /
• pp.8-13
• /
• 2014

In this paper, the gait imbalance evaluation algorithm based on temporal symmetry ratio using encoder is proposed. The device is attached to the hip joint in order to measure the angle during the normal gait. Using an angle data, the stance phase and swing phase was determined. And the value of TSR(temporal symmetry ratio) was calculated by stance phase and swing phase of gait cycle. For the comparative experiment, the conventional method of the foot pressure was measured at the same conditions. The results of statistical analysis, there was a significant difference (p < 0.05) when using encoder. The gait imbalance analysis using encoder is effective in determining the imbalance than using the existing method of pressure.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.5A
• /
• pp.341-346
• /
• 2009

M-WiMAX (Mobile-Worldwide Interoperability for Microwave Access) system, which uses OFDM (Orthogonal Frequency Division Multiplexing) technique, is known to be proper for mobile high-speed data transmission system. Nevertheless, M-WiMAX is seriously sensitive to IQ imbalance caused by the LO (Local Oscillator) at the receiver. In this paper, we analyze the effect of IQ imbalance on the system, and then propose a joint optimization scheme that can optimize DFE (Digital Front-end) of mobile modem by combining operation duplicated between AGC (Automatic Gain Control) and the estimation and compensation of IQ imbalance. Simulation results show that the proposed scheme achieves the same performance of the conventional scheme while reducing the complexity of the H/W implementation.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.7
• /
• pp.671-679
• /
• 2021

House ownership is considered as one of the important pre-conditions for marriage in China. Given that gender imbalance is a prominent issue in the country, competition for marriage partners might motivate males to look for a house and probably bigger and more expensive house. This is believed to have caused house price hikes in recent years. This study aims to investigate the impact of gender imbalance on house prices using data from 30 provinces in China for the 2000-2017 period. The results based on the generalized method of moments (GMM) estimations show that house price is strongly influenced by gender imbalance. However, there is no evidence to support differential effects across eastern and mid-western regions. One potential reason is that pre-marriage house ownership has become a common culture for the whole community and therefore it does not vary significantly across regions. There are several important policy implications. Firstly, the issues should be addressed by the policymakers at national level and not regional level. Secondly, the government should intervene to bring back gender ratio to its normal level. Finally, the government should limit the number of houses people can buy and increase the supply of houses in the market.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.3
• /
• pp.25-33
• /
• 2021

Energy-harvesting wireless sensor networks(EH-WSNs) can collect energy from the environment and overcome the technical limitations of existing power. Since the transmission distance in a wireless sensor network is limited, the data are delivered to the destination node through multi-hop routing. In EH-WSNs, the routing protocol should consider the power situations of nodes, which is determined by the remaining power and energy-harvesting rate. In addition, in applications such as environmental monitoring, when there are urgent data, the routing protocol should be able to transmit it stably and quickly. This paper proposes an adaptive routing protocol that satisfies different requirements of normal and urgent data. To extend network lifetime, the proposed routing protocol reduces power imbalance for normal data and also minimizes transmission latency by controlling the transmission power for urgent data. Simulation results show that the proposed adaptive routing can improve network lifetime by mitigating the power imbalance and greatly reduce the transmission delay of urgent data.