Altmetrics are indicators about the impacts of scientific research that are calculated using data extracted from the social Web. For example, counting the number of times that an article had been tweeted about or blogged about in Sina Weibo might be used as an altmetric indicator of interest in the article. This perspective argues that altmetrics are very suitable for data science research because they are relatively easy to collect on a large scale, have non-trivial statistical properties, and can be used to help in the important issue of developing methods to understand and assess the impact of scientific research.