The following visualizations were constructed using a collection of over two million New York Times articles. They aim to illustrate common patterns, be that cultural or semantic, that exist in NYT articles. While we do not have comprehensive datasets for other news agencies to compare to, we aim to extrapolate our findings as an explanation to broader news trends.
This visualization, called a Treemap, shows the breakdown of published articles by section for a given year. Treemaps are effective at conveying how individual sections contribute to the total volume of published articles. By interacting with the year selector, users can explore trends over time. For example, they might discover growth in technology-related articles with the rise of the internet and digital culture. They might also discover the New York section was common during the post-9/11 period, that the Business Day section rose during the 2007 financial crisis, and that the World section rose during the 2003 Iraq War. Viewers may also notice that some sections hold a relatively constant percentage makeup of total articles published, while others fluctuate greatly year to year.
"Hot topic" words can give us insights into the current events of a given year. This word cloud shows for each year the top ten keywords that saw the largest percentage increase in frequency of use from the previous year. We decided on a word cloud because it allows users to visualize the growth of top keywords and compare them relative to others. The size of each word indicates the magnitude of growth, making it easy to identify which topics have gained the most traction. As one might expect, in 2001, the keywords 'New York City' and 'Terrorism' grew significantly from the previous year, reflecting the impact of the September 11th attacks on public consciousness and media coverage. In 2003, we see the prominence of keywords like 'Iraq', 'United States International Relations', and 'Housing', coinciding with the Iraq War and the focus on real estate that foreshadowed the coming housing market crisis. In 2008, 'Barack Obama' and 'United States Economy' surge in frequency, corresponding to Obama's election campaign and the global financial crisis.
This line chart illustrates the average sentiment across various sections over time. We utilized sentiment analysis to gauge the emotional tone of the articles in each section, presenting a clear picture of how the tone of news changes. This visualization shows how different sections of news might respond to the same event. For example, the September 11 attacks in 2001 and the financial crisis of 2008 led to significant shifts in sentiment across all sections, highlighting how major events have a widespread impact on reporting.
Our visualizations have provided valuable insights into the publishing trends, keyword growth, and sentiment dynamics of New York Times articles over the years. By exploring the data through these visualizations, we can better understand how The New York Times has responded to historical events, how public interests have shifted, and how the tone of news reporting has evolved. We hope these visualizations inspire further exploration and analysis of media trends. We would like to thank our professor, Bret Jackson, for teaching us the visualization tools and techniques needed for this project.