Roy Huang
About Roy Huang
Roy Huang is a Data Analyst Intern with extensive experience in gene feature selection and classification models. He has worked at various organizations, including mProbe Inc. and eBizTie Inc., while pursuing a Bachelor's degree in Computer Science at the University of California, Santa Cruz.
Work at mProbe Inc.
Roy Huang has been employed as a Data Analyst Intern at mProbe Inc. since 2018. In this role, he has contributed to various data analysis projects, utilizing advanced statistical methods and machine learning techniques. His work includes conducting analyses related to stem cell activity and classification models, demonstrating his proficiency in data-driven decision-making.
Education and Expertise
Roy Huang is pursuing a Bachelor's degree in Computer Science at the University of California, Santa Cruz, where he has been a student since 2015. His academic focus includes data analysis, machine learning, and bioinformatics. He has developed expertise in statistical analysis and programming, applying these skills in both academic and internship settings.
Background in Data Analysis
As part of his internship experience, Roy conducted an analysis of intestinal injury-inducible stem cell activity. He selected approximately 200 significant gene features from a dataset of 14,000 using t-tests and fold change. Additionally, he performed supervised learning feature selection for microarray prediction analysis, showcasing his ability to handle complex datasets.
Previous Experience at eBizTie Inc.
Prior to his current role, Roy Huang worked as a Mobile & Web Developer Intern at eBizTie Inc. in San Jose for seven months in 2018. This experience provided him with foundational skills in software development and project management, contributing to his overall technical proficiency.
Achievements in Data Classification Models
Roy developed a two-label Infantile Asthma KD/FC classification model using Keras in R, achieving a testing accuracy of 0.95 and an AUC of 0.945 in ROC tests. He also utilized Random Forest (Boruta) to select significant features for classification tasks, demonstrating his capability in applying machine learning algorithms to real-world problems.