ABSTRACT
Social media is becoming an increasingly important tool for journalists to find news content and to distribute the stories to their audience. Social media often releases news about current affairs that are trending, which the mainstream media might not have been aware of. The newsrooms hence rely on social media feeds to broadcast and publish the news. For the media to track news, hashtags are used. The term “Hashtag” allows users to group messages from the Internet and extract information that pertained to a specific theme of context. The group of words can be combined and prefixed by a hashtag. Hashtags contextualize the entire story in minimum words, which helps in keeping track of the news. In this work, we propose a model to find the most suitable hashtag for news articles and tweets using a probabilistic approach. The proposed probabilistic neural network (PNN) model provides 264the probability metric for each hashtag that can be appropriate for a specific event. As a result, for data segmentation, this model can integrate news and social-stream in real time and further processing. The proposed model can deliver recommendations and narrate the incidents that happened before and after the current hot event. This can aid the media industry to streamline its delivery. The recommended hashtags can be used by the journalist for associating the news stream with the social stream. This strong association can address challenging real-time issues and can capture the dynamic evolution of news. Our model is implemented using news articles from RSS feeds and tweets from Twitter Streaming API.
