It will collaborate with Singapore research agency A*STAR’s Institute of High Performance Computing (“IHPC”) to first identify elements of news headlines that make an article popular, using IHPC’s proprietary technology for sentiment and emotion analysis.
Both parties will then develop a system that conducts affective analysis to predict the virality of an article based on its headline. This will enable SPH journalists to test different headlines for their articles, before choosing the one that would be most effective in capturing a reader’s interest and attention.
For this endeavour, IHPC will work with SPH’s data science team from the company’s Media Strategy and Analytics department, which oversees efforts to use data analytics and artificial intelligence to improve on SPH’s media products. Three months ago, SPH announced its partnership with DC Frontiers to develop a content recommendation engine powered by machine learning.
“This project will help our newsrooms better understand the emotional impact of different words and phrases used in headlines, so we can improve our engagement with readers and guide them towards important and informative news without resorting to clickbait,” said Anthony Tan, Deputy CEO of SPH.
In a concurrent press announcement, SPH also unveiled its new stock photo marketplace called Photonico.
This marketplace is focussed on curating Asian contemporary culture and heritage photos that will be organised into categories including Singapore Skyline, Singapore Heritage, Daily Life, Food, Nature and Animals, Business and Asian Countries.
Also Read: Singapore’s Basil Partners closes its third fund, focusing on digital tech service
Currently, the marketplace has an archive of 12,000 photos, with some dating back to 1948. More are expected to be added to the repository every day. Photonico is currently working with a small group of 6 photo contributors at the moment but willgradually scale it up.
In future iterations of the website, external photo contributors will be able to upload and contribute photos and earn a revenue share from showcasing and selling their portfolio of photos.
Image Credit: SPH