You are here: Home » International » News » Technology
Business Standard

Robust Label Ranking can help predict emoticons on Facebook posts

ROAR can help navigate the increasingly complicated way people are expressing how they feel on social media

IANS  |  New York 

Facebook Emotions

A novel computer model called robust label ranking, or ROAR, that helps mine social emotions, could be used to predict various emotional reactions such as love, haha, and angry on posts, researchers say.

While the trusty "like" button is still the most popular way to signal approval for posts, can help navigate the increasingly complicated way people are expressing how they feel on

It could also lead to better analytic packages for analysts and researchers.

"We want to understand the user's reactions behind these clicks on the emoticons by modelling the problem as the ranking problem -- given a post, can an algorithm predict the right ordering among six emoticons in terms of votes?" said Jason Zhang, a research assistant at the

"This is a step in the direction of creating a model that could tell, for instance, that a posting made in 2015 with a million likes, in fact, consists only 80 per cent likes and 20 per cent angry," Lee said.

In early 2017, added five more buttons -- love, haha, wow, sad and angry -- in addition to its the like button -- the official emoticon reaction.

But, merely counting clicks fail to acknowledge that some emoticons are less likely to be clicked than others.

For example, users tend to click the like button the most because it signals a positive interaction and it is also the default emoticon on

For managers and advertisers, who spend billions buying advertisements each year, this imbalance may skew their analysis on how their content is actually performing on Facebook, said Dongwon Lee, Associate Professor at the varsity.

The new model was trained using four post data sets including public posts from ordinary users, the New York Times, and

The findings, presented at the 32nd AAAI Conference on Artificial Intelligence on Tuesday in New Orleans, showed that the new model significantly outperformed existing solutions in precisely understanding

First Published: Wed, February 07 2018. 17:57 IST
RECOMMENDED FOR YOU