itle: “SwiftKey Exploratory Data Analysis”

uthor: “Swasthik Kotian”

ate: “2026-09-22”

utput: html_document

Introduction

The goal of this analysis is to explore the SwiftKey text data and understand its basic characteristics. The dataset contains three sources of text: blogs, news, and Twitter. The analysis focuses on the number of lines and words in each dataset and examines the distribution of words per line.

Basic Summary of the Data

The three datasets contain a large amount of text.

summary_table

##   Dataset   Lines    Words
## 1   Blogs  899288 37334131
## 2    News 1010206 34371031
## 3 Twitter 2360148 30373583

The blogs dataset contains 899,288 lines and approximately 37.3 million words. The news dataset contains 1,010,206 lines and approximately 34.4 million words. The Twitter dataset contains 2,360,148 lines and approximately 30.4 million words.

These results show that Twitter contains the largest number of lines, while the blogs dataset contains the largest number of words.

Words Per Line

To understand the structure of the text, a random sample of 10,000 lines was taken from each dataset. Histograms were created to examine the number of words per line.

The distributions show that most lines contain a relatively small number of words, while longer lines occur less frequently. Twitter has a different distribution from blogs and news, reflecting the shorter and more concise nature of social media messages.

Findings

Several useful findings were identified during the exploratory analysis:

• The three datasets contain millions of lines and words.
• Twitter has the largest number of lines.
• Blogs contain the largest total number of words.
• The number of words per line varies across the three data sources.
• Most observations contain relatively few words, while very long observations are less common.
• The data provides a useful foundation for developing a predictive text algorithm.

Future Plan

The next stage of the project will focus on building a prediction algorithm that can suggest the next word based on previously entered words. The analysis will use the text data to identify common word combinations and patterns.

A Shiny application will eventually provide an interactive interface where a user can enter text and receive a predicted next word.

Conclusion

The exploratory analysis confirms that the SwiftKey dataset has been successfully downloaded and loaded. Basic summaries and visualizations have been created to understand the size and structure of the data. These findings provide a foundation for developing the prediction algorithm and Shiny application.