The main goal of the capstone project is to build well-performing Shiny application, which predicts next word in sentence. Supported language is English. This presentation summaries our endeavour to the final data product.

This milestone report deals with investigation of en_US part of SwiftKey Corpora, as one step in Data Science Capstone course by John Hopkins University. The ultimate goal of the course is to design and implement predictive text model on ShinyApp infrastructure. The report aims to get familiar with the corpora and fundamentals of NLP, to report basic summaries, to investigate properties of the data with vizualization and to write about whole experience concisely.

The goal of this case study is to analyze impact of extreme weather events on US economics and population health. Source data are obtained from US National Oceanic and Atmospheric Administration (NOAA). Study uses NOAAs estimates of USD damage to properties and crops and estimates of fatalities and injuries count to identify the most severe weather events. The most damaging events with regard to US economy are flood, hurricane and draught. When we look at population health, the mosth hazadous events are tornado, excessive heat and floods.