This presentation is created as required by data science capstone - final project

The main aim of this project is to build natural language processing model which can predict next word in a sentence as accuratley as possible

• Data is obtained by coursera data science capstone project website (Swiftkey)
• Used tidytext package to tokenise data
• ngram method is used in the prediction of next words
• this app uses filtering words from 4 gram tokens

• The shiny app can be accessed by this link https://sandeepsan.shinyapps.io/final_app/

Cons

• The app uses only the 5% of the swiftkey data
• Only the lower case letters are detected

• tidytext package https://www.tidytextmining.com/
• Data Science Specialization (Coursera) - https://www.coursera.org/specializations/jhu-data-science