KOINEU

Finding the Sweet Spot: Optimal Data Augmentation Ratio for Imbalanced Credit Scoring Using ADASYN

1. 연구 배경 및 필요성 클래스 불균형 : 신용점수 모델은 디폴트(불량) 사례가 전체의 5~10% 수준으로 매우 희소하다. 이로 인해 학습 알고리즘이 다수 클래스에 편향되고, 소수 클래스(디폴트) 예측력이 저하된다. 기존 접근법의 한계 : SMOTE, ADASYN 등 합성 오버샘플링 기법은 소수 클래스를 인위적으로 늘려 균형을 맞추지만, “얼마나 많이” 증강해야 하는지에 대한 실증적 근거가 부족하다. 실무에서는 흔히 1:1(완전 균형) 비율을 무조건 적용한다. 2. 데이터 및 실험 설계 | 항목 | 내용 | | | | | 데이터셋

Finding the Sweet Spot: Optimal Data Augmentation Ratio for Imbalanced Credit Scoring Using ADASYN

FinSight: Towards Real-World Financial Deep Research

Focal Modulation and Bidirectional Feature Fusion Network for Medical Image Segmentation

Foundation Models for Medical Imaging: Status, Challenges, and Directions

Fourier analysis of the physics of transfer learning for data-driven subgrid-scale models of ocean turbulence

FPGA Innovation Research in the Netherlands: Present Landscape and Future Outlook

Fractional neural attention for efficient multiscale sequence processing

Frobenius method for Mahler equations

From Measurement to Expertise: Empathetic Expert Adapters for Context-Based Empathy in Conversational AI Agents

Gabor-Enhanced Physics-Informed Neural Networks for Fast Simulations of Acoustic Wavefields

GATMesh: Clock Mesh Timing Analysis using Graph Neural Networks

Gauge symmetry and the arrow of time: How to count what counts

General Strong Bound on the Uncrossed Number which is Tight for the Edge Crossing Number

Generalizable Reasoning through Compositional Energy Minimization

Generating Fair Consensus Statements with Social Choice on Token-Level MDPs

Generating Sizing Fields for Mesh Generation via GCN-based Simplification of Adaptive Background Grids

Generative Machine Learning in Adaptive Control of Dynamic Manufacturing Processes: A Review

GestureLSM: Latent Shortcut based Co-Speech Gesture Generation with Spatial-Temporal Modeling

Graph Unlearning Meets Influence-aware Negative Preference Optimization

Graph-Coarsening Approach for the Capacitated Vehicle Routing Problem with Time Windows

Graph-Enhanced Policy Optimization in LLM Agent Training

Ground state energy of the dilute Bose-Hubbard gas on Bravais lattices

Growing Mathlib: maintenance of a large scale mathematical library

Handwritten Text Recognition of Historical Manuscripts Using Transformer-Based Models

HeatGen: A Guided Diffusion Framework for Multiphysics Heat Sink Design Optimization

HERO: Hardware-Efficient RL-based Optimization Framework for NeRF Quantization

How News Feels: Understanding Affective Bias in Multilingual Headlines for Human-Centered Media Design

How to Auto-optimize Prompts for Domain Tasks? Adaptive Prompting and Reasoning through Evolutionary Domain Knowledge Adaptation

Hybrid DeepONet Surrogates for Multiphase Flow in Porous Media

IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks

Idefix-Closed Languages and Their Application in Contextual Grammars

Implementing Keyword Spotting on the MCUX947 Microcontroller with Integrated NPU

Implicature in Interaction: Understanding Implicature Improves Alignment in Human-LLM Interaction

Improving VANET Simulation Channel Model in an Urban Environment via Calibration Using Real-World Communication Data

Interaction Dynamics as a Reward Signal for LLMs

Interpretable Recognition of Cognitive Distortions in Natural Language Texts

Investigating Thinking Behaviours of Reasoning-Based Language Models for Social Bias Mitigation

Is string field theory background independent?

It Takes Two: A Dual Stage Approach for Terminology-Aware Translation

Knowledge and Common Knowledge of Strategies

Knowledge-based anomaly detection for identifying network-induced shape artifacts

KPZ-like transport in long-range interacting spin chains proximate to integrability

Languages of Words of Low Automatic Complexity Are Hard to Compute

Large Language Models (LLMs) for Electronic Design Automation (EDA)

Laser Scan Path Design for Controlled Microstructure in Additive Manufacturing with Integrated Reduced-Order Phase-Field Modeling and Deep Reinforcement Learning

Latent Diffusion Model without Variational Autoencoder

Lattice XBAR Filters in Thin-Film Lithium Niobate

Laugh, Relate, Engage: Stylized Comment Generation for Short Videos

Learning electromagnetic fields based on finite element basis functions

Learning Explainable Stock Predictions with Tweets Using Mixture of Experts

< Category Statistics (Total: 5461) >

Start searching

No results found