Abstract Handling imbalanced datasets in binary classification, especially in employment big data, is challenging. Traditional methods like oversampling and undersampling have limitations. This paper integrates TabNet and Generative Adversarial Networks (GANs) to address class imbalance. The generator creates synthetic samples for the minority class, and the discriminator, using TabNet, ensures authenticity. Evaluations on benchmark datasets show significant improvements in accuracy, precision, recall, and F1-score for the minority class, outperforming traditional methods. This integration offers a robust solution for imbalanced datasets in employment big data, leading to fairer and more effective predictive models.

본 연구는 경기도 베이비붐세대와 에코세대의 세대별 주거만족도 영향요인을 분석함으로써, 경기도 세 대별 주거만족도 향상을 위한 정책적 함의를 제시하는 것이 목적이다. 이를 위해서 ‘2020 경기도민 삶의 질 조사’ 자료를 활용하여 이분형 로짓분석을 적용한 실증분석을 진행하였다. 분석결과, 베이비붐세대(고령 세대)가 에코세대(청년세대)보다 주거만족도가 높았으며, 베이비붐세대의 경우 경기북부지역보다 경기남부 지역에서 주거만족도(공간적 분포)가 높았다. 청년세대인 에코세대는 전·월세에 거주하는 경우 주거만족도 가 베이비붐세대보다 높았고, 고령세대인 베이비붐세대는 1인가구일수록 주거만족도가 높았다. 이런 결과 를 봤을 때, 베이비붐세대와 에코세대의 세대별 주거만족도 향상을 위한 정책적 함의는 다음과 같다. 첫째, 베이비붐세대의 경우 1인가구(소형주택) 중심의 주택공급 전략이 필요하다. 둘째, 베이비붐세대의 주거안정 을 위해서는 주택공급의 다양화가 필요하다. 셋째, 에코세대의 주거안정을 위해서는 임대주택 중심의 주택 공급이 필요하다. 넷째, 에코세대를 타깃으로 한 주거비지원과 금융지원이 필요하다. 마지막으로, 에코세대 의 주거만족도 향상을 위해서는 지역별 차별화 전략이 필요하다. 본 연구는 세대별 주택정책 수립을 통해 경기도민의 주거복지를 향상시키는 데 기초자료로 활용될 것으로 기대된다.

The purpose of this study was to examine the extent to which influencing factors impact the transference willingness of farmers. The research method is based on the survey data on transference willingness, and the binary logistic model was used to analyze the influencing factors. The results show that the educational level, annual household income, farm labor, and farmers' understanding of land transfer policy notably influence land transfer. Based on the results, suggestions include improving farmers' social security system, improving a sound market for rural land transfer, and deepening the promotion and explanation of land-related policies.

We present the photometric properties of V608 Cas from detailed studies of light curves and eclipse timings. The light curve synthesis indicates that the eclipsing pair is an overcontact binary with parameters of ΔT = 155 K, q = 0.328, and f = 26%. We detected the third light ℓ3, which corresponds to about 8% and 5% of the total systemic light in V and R bands, respectively. Including our 6 timing measurements, a total of 38 times of minimum light were used for a period study. It was found that the orbital period of V608 Cas has varied in some combination of an upward parabola and two periodic variations. The continuous period increase with a rate of +3.99 × 10−7 d yr−1 can be interpreted as a mass transfer from the secondary component to the primary star at a rate of 1.51 × 10−7 M⊙ yr−1. The periods and semi-amplitudes of the two periodic variations are about P3 = 16.0 yr and P4 = 26.3 yr, and K3 = 0.0341 d and K4 = 0.0305 d, respectively. The most likely explanation of both cycles is a pair of light-traveling time effects operated by the possible presence of third and fourth components with estimated masses of M3 = 2.20 M⊙ and M4 = 1.27 M⊙ in eccentric orbits of e3 = 0.66 and e4 = 0.52. Because the contribution of ℓ3 is very low compared to the estimated masses of two circumbinary objects, they can be inferred as very faint compact objects.

The activation energy to create a phase transformation or for the reaction to move to the next stage in the milling process can be calculated from the slop of the DSC plot, obtained at the various heating rates for mechanically activated Al-Ni alloy systems by using Kissinger's equation. The mechanically activated material has been called “the driven material” as it creates new phases or intermetallic compounds of AlNi in Al-Ni alloy systems. The reaction time for phase transformation by milling can be calculated using the activation energy obtained from the above mentioned method and from the real required energy. The real required energy (activation energy) could be calculated by subtracting the loss energy from the total input energy (calculated input energy from electric motor). The loss energy and real required energy divided by the reaction time are considered the “metabolic energy” and “the effective input energy”, respectively. The milling time for phase transformation at other Al-Co alloy systems from the calculated data of Al-Ni systems can be predicted accordingly.