Prediction App

The prediction app algorithm works in the following way: -It takes the input text, cleans it, splits the words and counts them.

-Depending on the word count, it outputs the next predicted word using the ngrams.

-For example, if there is one word in the text input then, the next word is predicted using the bigram data. If there are two words then, the next word is predicted using the trigram data and so on.

-Ngrams upto 6 words, that is upto sextgram, have been used in this model.

-The training dataset used to create this app can be limited due to which a message for the same is displayed for less frequent words/ phrases.

-Enter a phrase or a word in the given input textbox in the sidebar.

-The app predicts the next word based on the input.

-Read about the algorithm and instructions to use the app in the tab panel named “About”.

-When the input is cleared the app displays a message to enter a text in the input box.

-If less frequent words are used, the app displays a message that the training set is limited.

The link to the shiny app can be found at https://aryava.shinyapps.io/Prediction_App/

The link to the source code can be found at https://github.com/AryaVa/Capstone-project

The milestone report can be found at http://rpubs.com/AryaVa/Milestone_report