洪慧念

A Statistical Analysis of Student Cross-Border Mobility in Taiwanese Higher Education Based on Public Institutional Data from Taiwan's Ministry of Education, Academic Years 106-113

統計學研究所・共同指導：王維菁

校務資料分析、學生跨境流動、出國交流、雙聯學位、群組化比例迴歸模型、群集分析、institutional research、student cross-border mobility、student exchange、dual-degree programs、grouped proportion regression model、cluster analysis

Propensity Score Design Diagnostics for Causal Inference with Survival Outcomes

統計學研究所・共同指導：王維菁

因果推論、傾向分數、觀察性研究、限制平均存活時間、逆機率加權、共變數平衡、機器學習、causal inference、propensity score、observational studies、restricted mean survival time、inverse probability weighting、covariate balance、machine learning

A Study on Information Gap Bridging in Wearable ECG Devices for Early Classification

數據科學與工程研究所・共同指導：曾新穆

穿戴式裝置、心電圖分類、訊號重建、知識蒸餾、及早序列分類、深度學習、Wearable Devices、ECG Classification、Signal Reconstruction、Knowledge Distillation、Deep Learning、Early Time Series Classification (ETSC)

Reconstruction-based Aanomaly Predictor for Exetremely Imbalanced Multivariate Time-series Anomaly Prediction

數據科學與工程研究所・共同指導：曾新穆

異常預測、異常偵測、不平衡、自編碼機、剩餘使用壽命預測、Anomaly prediction、Anomaly detection、Imbalance、Autoencoder、Remaining useful life prediction

Deep Survival Framework Based on C-index

統計學研究所・共同指導：王維菁

生存分析、深度學習、C指數、粒子群最佳化、基因演算法、個體化風險預測、Survival analysis、deep learning、C-index、metaheuristic optimization、particle swarm optimization、hyperparameter tuning、censored data

Observations on the 2019 Curriculum Reform and Grade Inflation: Analysis Based on NYCU Administrative Data

統計學研究所・共同指導：王維菁

108課綱、分數膨脹、基礎物理、微積分、108 Curriculum、Grade inflation、Physics、Calculus

DiPPSI : Diffusion-based Pulsative Physiological Signal Imputation

數據科學與工程研究所・共同指導：曾新穆

行動健康、三軸加速度計、條件擴散模型、信號補值、脈衝模式、mHealth、accelerometers、diffusion model、signal imputation、pulse pattern

Statistical Analysis of Unreported COVID-19 Cases

統計學研究所・共同指導：王維菁

新冠肺炎、未通報人數、COVID-19、unreported cases

Prediction of Ariake Seawater Depth in Japan Using Multi-Resolution Spline Basis Functions and Truncated Gaussian Processes

統計學研究所・共同指導：黃信誠

薄板樣條基底函數、高斯過程、截斷多元常態分佈、簡單克里金法、Thin plate spline、Gaussian processes、truncated multivariate normal distributions、simple kriging

Spatial Sparse Principal Component Analysis Using Autoencoders with Convolutions

統計學研究所・共同指導：黃信誠

典型相關分析、類神經網路、非線性主成分分析、正則化、Canonical correlation analysis、Neural network、Nonlinear principal component analysis、Regularization

Privacy-Preserving Attention-Weighted Multi-Source Domain Adaptation for EEG Motor Imagery

數據科學與工程研究所・共同指導：曾新穆

動作意象任務、多源域權重、隱私保護、非監督式域適應、注意力機制、Motor imagery、Weighted multi-source、Privacy-preserving、Unsupervised domain adaptation、Attention mechanism

A Comparison of LSTM and Traditional Time Series Models

統計學研究所

時間序列、深度學習、長短期記憶、機器學習、降雨量預測、Time Series、Deep Learning、LSTM、Machine Learning、Rainfall Prediction

A Prediction Model of Mixed Martial Arts

數據科學與工程研究所・共同指導：王維菁

綜合格鬥、深度學習、圖、重採樣、勝負預測、mixed martial arts、deep learning、graph、bootstrap method、win-loss prediction

Early Multi-label Classification for Long-length and Varied-length Time Series

數據科學與工程研究所・共同指導：曾新穆

早期分類、多標籤分類、時間序列分類、深度強化學習、注意力機制、Early Classification、Multi-label Classification、Time Series Classification、Deep Reinforcement Learning、Attention Mechanism

Analyzing the Behaviors of NCTU Students in Double Minor/Major, Dropout, and Course Withdrawal

統計學研究所・共同指導：王維菁

輔系、雙主修、學業表現、隨機效應模型、退學制度、停修制度、Minor majors、Double majors、Academic performace、Random-effects model、Dropout policy、Course withdrawal policy

The Impact of Student Background and Academic Performance on College Admission

統計學研究所・共同指導：王維菁

多元入學政策、城鄉差距、家戶所得、學生背景、個人申請第二階段、考生特質、Multiple Entrance Program Policy、Rural-urban gap、economic background、students' background、second stage of individual application、Candidate Characteristics

Reconstructing Paleoclimate in China Using REACHES Data

統計學研究所・共同指導：黃信誠

古氣候、簡單克利金法、夏季季風指標、資料同化、卡爾曼濾波器、Paleoclimate、simple kriging、summer monsoon index、data assimilation、Kalman filter

Integrated Interictal Epileptiform Discharges Detection from Scalp EEG Signals: Methods and Evaluation

數據科學與工程研究所・共同指導：曾新穆

癲癇波偵測、頭皮腦波圖、深度學習、規則法、模板比對、Iterictal Epileptiform Discharges Detection、Scalp Electroencephalography、Deep Learning、Rule-based、Template Matching

Multi-Step Predictions for Commodity Prices with High-Dimensional Mixed-Frequency Exogenous Information

數據科學與工程研究所・共同指導：曾新穆

原物料價格預測、經濟指標、巨量資料、混和頻率時間序列、深度學習、多步預測、commodity prices prediction、economic indicators、big data、mixed-frequency time series、neural network、multi-step forecasting

2021 MLB Tiebreaker Analysis with Markov Model

統計學研究所・共同指導：王維菁

突破僵局制、馬可夫模型、轉移矩陣、得分分佈、勝負預測、tiebreaker rule、Markov model、transition matrix、run distribution、win-loss prediction

Model Architecture Design and Parameter Estimation for Few-shot Learning

數據科學與工程研究所

自動化機器學習、小樣本學習、神經網路架構搜尋、Automated machine learning、Few-shot learning、Neural Architecture Search

Early Time Series Classification for Anomaly Prediction

數據科學與工程研究所・共同指導：曾新穆

時間序列、早期分類、異常預測、預測性維護、Time series、Early classification、Anomaly prediction、Predictive maintenance

A Statistical Optimization Algorithm for Deep Learning

統計學研究所

優化演算法、模擬退火演算法、深度學習、神經網路、統計計算、optimization algorithm、simulated annealing algorithm、deep learning、neural network、statistical computing

High Dimensional Bayesian Optimization Via Narrow Down Parameter Space

統計學研究所

貝式最佳化、調教超參數、高維度的、機器學習、bayesian optimization、hyperparameters tuning、high dimensional、machine learning

The Decomposite T2 -Test Under Large Dimension

統計學研究所・共同指導：陳志榮

複合式 T 檢定、高維度下共變異數矩陣、旋轉不變性估計值、Decomposite T test、high-dimensional covariance matrix、rotation-equivariant estimator

The Institutional Research – policy in homogeneity in senior high schools and students’ performance in calculus.

統計學研究所

多元入學政策、學業表現、multi-route admission program、academic performance

Soft-Timing Impact for Soft-PHY Enhancement

資訊科學與工程研究所・共同指導：許騰尹

第五代行動網路、軟體化PHY優化、時序化重新建構、LTE、Soft-Timing、MAC layer、Memory swapping

Comparison of Feature Selection Methods in Statistics and Machine Learning

統計學研究所

變數選取、逐步迴歸、機器學習、Feature Selection、Stepwise Regression、Machine Learning

A Statistical Model for Assessing Different University Admission Channels

統計學研究所・共同指導：王維菁

入學管道、申請入學、指考分發、數學模型、Admission by Application、Assigned-Subject Examination、Admission Channels、Statistical Models

A review of the Bayesian division estimation of the multidimensional density function

統計學研究所

貝氏、多維度、密度函數估計、重要性抽樣、Bayesian、Multidimensional、Estimation of the density function、Importance Sampling

Application of Machine Learning in Financial Return Anomalies

統計學研究所・共同指導：黃宜侯

異常報酬、機器學習、投資組合、特徵轉換、Anomalies、Machine Learning、Portfolio、Feature Transform

Radial Basis Function Neural Network for Losses of Rice

統計學研究所

輻狀基底類神經網路、稻米損失、Radial basis function neural network、Losses of rice

Application of Multivariate Random Effect Model to Multivariate Meta-analysis and Multiregional Clinical Trials with Multiple Endpoints

統計學研究所・共同指導：蕭金福

整合分析、多區域臨床試驗、多維度t分配、Meta-analysis、MRCT、Multivariate t distribution

The Estimation of Restricted Multinomial Distribution Parameters

統計學研究所

限制型多項分佈、Restricted Multinomial Distribution

A Dynamic Weighting Method and Analysis

統計學研究所

加權蒙地卡羅法、蒙地卡羅法、Monte Carlo Markov Chani、Dynamic Weighting、Importance sampling、Metropolis-Hasting Algorithm

Variable selection methods

統計學研究所

變數選取、P大於N、Stochastic search variable selection、Bayesian Lasso、large p and small n、Variable selection

The Analysis of Stochastic Volatility in High-Frequency Financial Data

統計學研究所

高頻率財務資料、波動率、隨機模型、時間序列模型、高維度資料、High-Frequency、Volatility、Stochastic Model、Time Series Model、High-Dimension

Students’ Background and Academic Performance in the second year under Multi-route Admission Program based on NCTU Students

統計學研究所

多元入學政策、學生背景特質、學業表現、multi-route admission program、students’ background、academic performance、the best admission strategy

Evaluating the Effectiveness of Random Forest Model

統計學研究所

隨機森林、Random Forest

Review of AIC, BIC and EBIC

統計學研究所

高維度模型、模型選取、AIC、BIC、EBIC、High Dimensional Model、Model Selection

Bayesian approach for group supplier selection based on the process capability indices

統計學研究所・共同指導：洪志真

供應鏈管理、群體供應商選擇、貝氏方法、事先資訊、正確選擇的後驗概率、Supply chain management、Group supplier selection、Bayesian approach、Prior information、Posterior probability of correct selection

Polya tree process

統計學研究所

Polya 樹先驗分配、貝氏無母數分析、隨機空間分割、Polya tree prior、Bayesian nonparametric、Randomised Partitions

LDV Model Parameter Estimater of Transaction Cost

統計學研究所

LDV模型、LDV Model

The Net Benefits to Leverage

統計學研究所・共同指導：吳儀玲

舉債淨效益、The Net Benefits to Leverage

Dirichlet Process and Chinese Restaurant Process

統計學研究所

隨機過程、狄里克雷過程、中國餐廳程序、Stochastic process、Dirichlet process、Chinese restaurant process