Article on unsupervised machine learning of materials science literature one of the top talked-about articles based on altmetric score

Vahe Tshitoyan recent published an article on unsupervised machine learning of materials science literature (Unsupervised word embeddings capture latent knowledge from materials science literature). Just one month later, that article is one of the most talked-about articles based on the “altmetric score”. The altmetric score is based on an algorithm that tries to quantify the amount of attention a journal article receives, so to speak. It generally does this by counting the number of times the article is mentioned or linked online, weighting mentions in higher-impact media such as the news and blogs more than mentions in tweets or reddit. Of the articles that are also scored using this algorithm, Tshitoyan’s article is the 8th highest ranked article in Nature of a similar age and the 54th highest ranked article in all journals of a similar age.

For more info, including blogs and news sites that linked the article, check out the article’s altmetric score here (article altmetric score). For a fun look at a public discussion on the Science subreddit (with >3k upvotes), check out the reddit post here (article reddit post).