My name is Luyao Peng. I am a senior research engineer at ByteDance with educational background in statistics (Ph.D.) and quantitative methodology in education (M.A.). My research interests are cheating behavior detection and adaptive testing in educational assessment and probabilistic deep learning models in NLP (natural language processing).

My current C.V.

My Journal Publications:

• Luyao Peng, Sandip Sinharay. 2020. The Use of Theory of Linear Mixed-Effects Models to Detect Fraudulent Erasures at an Aggregate Level. Educational and Psychological Measurement (Accepted)
• Subir Ghosh, Li Guo, Luyao Peng. 2018. Variance Component Estimators OPE, NOPE and AOPE in Linear Mixed-effects Models.Australian & New Zealand Journal of Statistics60(4), 481-505.
• Gregory Palardy, Luyao Peng. 2015. The Effects of Including Summer on Value-added Assessments of Teachers and Schools.Education Policy Analysis Archives, 23(92), 1-26.

My Personal Research Projects:

• Gaussian Process Classification by Tensorflow-probability (GPC jupyter notebook example)
• Bert Summarization and Generation of Chinese Text
• Using linear mixed-effects model to detect fraudulent erasures at an aggregate level (National Council on Measurement in Education 2020, conference abstract, pg113)
• An algorithmic construction of all unbiased estimators of variance components in linear mixed effects models (Joint Statistical Meetings 2019, conference abstract)

My R Packages:

• [MMeM] (https://cran.r-project.org/web/packages/MMeM/index.html)
• [regrrr] (https://cran.r-project.org/web/packages/regrrr/index.html)

My Visualization of Data Science Projects:

• Cheating behavior detection in essay scoring and forensic analysis (Intern Research Project at CTB, McGraw-Hill Education)
• Stock price visualization (milesone project at The Data Incubator)
• NLP essay grader (Capstone project at The Data Incubator)