What were the biggest headlines that captured the attention of readers over the years? And what’s the top news on the day you were born?
This was one of the projects that marked ST’s 175th anniversary in 2020.
The Straits Times’ digital graphics team captured and analysed more than 47,000 headlines that appeared on page one of The Straits Times from the very first edition on July 15, 1845 until October 15, last year, as well as those from The Sunday Times.
The interactive uncovered topics and stories in their respective years through data analysis methods that could have been difficult to discover using manual curation. It also allows readers to explore and recall the past using interesting insights.
For example, under “epidemics”, users will learn that while Covid-19 might be dominating the headlines today, Singapore has survived previous outbreaks of deadly diseases, such as smallpox in the 1950s. In 1959, ST reported that a 49-year-old was the first person to die of smallpox here.
The team grouped the headlines into five major topics: epidemics, foreign affairs, local towns, vices and crimes, and war and peace. Each category then allows the reader to browse through 5 to 7 topics highlighted by the respective headline images and click to find a description or story behind the trend or event.
The project uses an algorithm to help in the analysis. The algorithm gives weight to how important subjects were in their respective years. The charts help visualise how heavily each topic was mentioned in the headlines.
What was the hardest part of this project?
Surprisingly, gathering the headlines data. It is an old newspaper and it’s digital archives are not accessible to the publication through a machine-readable format. The team had to build a scraper that could pull first page content as far back as possible. Also, the front pages were typically used as classfieds or ads in the beginning, so we had to manually identify at which point The Straits Times changed editorial priorities for the front page.
What can others learn from this project?
For big datasets such as these, pulling out the interesting stories was fun and also important for highlighting them to the readers. Inspiring them to then want to explore the full data at the end has turned out to be a more engaging technique.