“Media bias” or specifically “social media bias” has been one of the major concerns of social media. The most popular social networking sites are Facebook and Twitter in the recent years. With the rise of social media, there has been a noticeable advent of social media bias mainly on these Web sites. To measure bias in social media, the betweenness and closeness centrality measure are calculated. More the betweenness centrality, more quickly the news spreads in the network is a general conclusion. In this paper, we assess the social media bias of Facebook and Twitter with the help of centrality measures and machine learning algorithms to find which one is more politically bias and to predict which algorithm can be best used to classify news as biased or unbiased. © 2021, Springer Nature Singapore Pte Ltd.