Summarizes a Shiny app for next-word prediction using an n-gram backoff model. It describes the algorithm, explains how the app works, gives user instructions, evaluates the user experience, and highlights the project’s value as a simple but functional NLP data product.

Simulated Supreme Court Data relating judicial retirement to tenure, workload, and presidential/senate alignment.

Voter registration by party affiliation broken down on a county basis.

Final Project for APD!

Estudo comparativo do desempenho do cursos de Medicina da UFPE no Enamed 2025.

Laporan ini menyajikan analisis komparatif menggunakan metode Linear Discriminant Analysis (LDA) dan Ordinal Logistic Regression (OLR) untuk mengklasifikasikan tingkat kebugaran berdasarkan Body Performance Dataset. Mencakup tahap preprocessing data, penanganan outlier, hingga evaluasi akurasi model.

Regresi Multinomial

This report presents a comprehensive analysis of website traffic and user behavior using a dataset of 2,000 user sessions. Key metrics include page views, session duration, bounce rate, time on page, previous visits, conversion rate, and traffic source (direct, organic, paid, referral, and social). The analysis covers data quality assessment, traffic source performance comparison, engagement metrics relationships, statistical testing (t-test, ANOVA, and regression), and visualizations (bar charts, scatter plots, histograms, boxplots, correlation matrix, and ECDF). Findings identify organic as the top traffic source by volume, referral as the most engaging by session duration, and repeat visitors as significantly higher converters. The report concludes with outlier detection and a multi-metric comparison across traffic sources.