Sentiment Polarity Identification in Banner Headlines of Broadsheets in the Philippines

Abstract

This study analyses the sentiment polarity of the banner headlines from six broadsheets in the Philippines with the biggest circulation nationwide. The sentiment polarity is the general perception of whether it is worded positively, neutral, or negatively. This study employs five machine learning and artificial intelligence (AI) to conduct the analysis. The results reveal a tone reflecting editorial policy that tends to lean towards the negative tone. While there is a utility for negative framing of the news, this paper argues that a pivot to the positive, particularly in the Philippine setting, is worth considering. Based on current literature, positivity shows the potential to bring in more readers. Publishers can leverage positivity in the news as part of strategies to stem the tide of readership decline. Positivity in the news should start with the headline, through which readers have a first glimpse of the story.

Keywords: sentiment polarity identification, sentiment analysis, sentiment polarity of banner headlines