This presentation pitches a data product designed for the Coursera Data Science Capstone. It outlines the development of a next-word prediction algorithm using N-gram modeling and a "Stupid Backoff" strategy. The project demonstrates the full data science pipeline: from exploratory analysis of a 4-million-line corpus (Twitter, News, Blogs) to the deployment of a reactive Shiny application. The focus is on balancing computational speed with predictive accuracy for mobile-first environments.

menjawab hipotesis 1 hingga 6

An interactive Plotly scatter plot created using R Markdown, visualizing the Iris dataset.

You can leave this blank or type "Course project for Developing Data Products

MPG Prediction Shiny Application

This presentation explains an interactive Shiny application that predicts vehicle fuel efficiency (MPG) based on car weight and horsepower using linear regression. It includes problem motivation, application overview, example prediction, and visualization.

Reproducible R code and analysis for a meta-analysis of menstrual disorders in Southeast Asian medical students. Includes data cleaning, random-effects modeling (using the 'meta' package), subgroup analysis, heterogeneity assessment (I2, prediction intervals), and high-resolution visualization (Forest plots, Funnel plots).

This presentation was created using R Markdown and includes an interactive Plotly visualization along with the document creation date.