PAF 516 Final Dashboard - Christopher Ogino

This chapter dives into the "engine room" of Data Science: Preprocessing. Students will learn that the quality of a model is determined long before it is trained, focusing on the critical steps required to turn messy, real-world data into a "model-ready" format. Core Topics covered: Why Preprocessing Matters Handling Missing Data Outlier Detection and Treatment Data Transformation Encoding Categorical Variables Feature Scaling Data Integration and Reshaping Chapter Lab Activity: Full Preprocessing Pipeline with msleep

This chapter addresses the foundational question of data science: "How do we ensure our data truly represents the world?" It explores the mechanics of selection, the math of sample size, and the power of computational resampling. Core Topics covered: Why Sampling Matters Probability Sampling Methods Non-Probability Sampling Methods Sample Size Determination Sampling Bias and Common Pitfalls Bootstrap Resampling Evaluating Sample Quality Chapter Lab Activity: Exploring Sampling with nhanes-Style Data

This chapter explores the statistical frameworks used to detect and quantify relationships between variables. It moves from testing the independence of categorical factors to measuring the strength and direction of associations in both discrete and continuous data. Core Topics covered: The Concept of Independence Chi-Square Test of Independence Fisher’s Exact Test Cramér’s V and Effect Size for Categorical Association Correlation Tests Point-Biserial and Phi Coefficients Partial Correlation Chapter Lab Activity: Exploring Independence with the titanic and mtcars Datasets

STAT 3280 Project

This report presents exploratory data analysis of the TMDB 5000 movies dataset. It includes data cleaning, visualization, and analysis of trends, relationships, and patterns such as ratings, revenue, budget, genres, and movie performance.

Final version of the colorectal cancer and EBV association if any, few genes found, we extracted, filtered, got the fold change values of knockout of AURKA gene in colon cancer commecial line SW480 and found PDL1 and PDL2 showed and MCC gene of EBV genes but not the others, then found all these genes predicted the class with 100% accuracy, and then added these genes to the pathology database, and added a link to a youtube playlist on this project and put together a partial dashboard of EBV associated pathologies with this data to add to it.