This project presents an algorithm model for predicting the next word in the Shiny application. From the SwiftKey database (news texts, blogs and Twitter), made available by Coursera during the Data Science Capstone course, a word prediction model was developed. For this, the data were cleaned (removal of special characters, punctuation, numbers and white spaces), tokenized and converted into n-grams of 1 to 4 words, considering the frequency of the words. Having the data classified by n-grams, a prediction model was elaborated, based on the back-off model.

This project presents an algorithm model for forecasting next word in the Shiny application, using the SwiftKey database, made available by Coursera during the Data Science Capstone course.

Geochron is a app to determine the age (in years) of geological samples. This app was created as a project during the Data Product course from Coursera.

The deforestation of the Amazon forest occurs for several reasons, such as illegal agriculture, natural disasters, urbanization, and mining, being frequent the occurence of burning and or wood extraction. This data set was obtained from a file “inpebrazilianamazonfires1999_2019”, from the National Institute for Space Research (INPE), which brings the number of firespot in the Brazilian Amazon by state, month and year, from 1999 to 2019. The original data are public and were extracted from the INPE website on December 13, 2019. Available at https://www.kaggle.com/mbogernetto/brazilian-amazon-rainforest-degradation?select=inpe_brazilian_amazon_fires_1999_2019.csv.

The deforestation of the Amazon forest occurs for several reasons, such as illegal agriculture, natural disasters, urbanization, and mining, being frequent the occurence of burning and or wood extraction. This data set was obtained from a file “inpebrazilianamazonfires1999_2019”, from the National Institute for Space Research (INPE), which brings the number of firespot in the Brazilian Amazon by state, month and year, from 1999 to 2019. The original data are public and were extracted from the INPE website on December 13, 2019. Available at https://www.kaggle.com/mbogernetto/brazilian-amazon-rainforest-degradation?select=inpe_brazilian_amazon_fires_1999_2019.csv.

The damage from severe weather events was based on the analysis of data available in the United States Oceanic and Atmospheric Administration (NOAA) storm database. For that, fatality, personal injury, property damage and damage to crops resulting from the events were considered as variables. From the interpretation of the graphs, tornado and heat showed the greatest health damage (fatalities and injury variables), being the most impactful for the population. In turn, thunderstorms, rains and storms presented the greatest damage to properties and crops, having the greatest economic consequences.