Abstract

In today's world, social media has evolved into a platform where various forms of content are shared daily. It is utilized by people with a wide range of viewpoints, beliefs, and motivations. Various topics are brought to light through social media at various times. Twitter has turned into a popular platform for discussing major events that are taking place throughout the world. It has enhanced the opportunity for audiences to generate news and freely share thoughts on news traversing through mass media. Although Twitter is a social media platform, in the Sri Lankan user context, it is widely used as a platform for news dissemination and journalism. Many recent incidents points to the assumption that social media platforms like Twitter can generate news-worthy pieces which are later discussed in mass media. A qualitative and quantitative analysis was conducted in this study on Twitter trending topics to understand how Sinhala Twitter data affects news dissemination on mass media. This study presents a set of features that adds a "news value" to a tweet, which can be utilized by the journalists or mass media to find news-worthy content from social media. A Neural Network model with the use of Clustering + Topic modeling for determining the topics shared in Sinhala is proposed as the solution which holds an F1 score of 0.73 and an accuracy of 64.41%.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call