A news app compares its AI-written article summaries with summaries written by its editors. It wants a fast metric that checks how much of the editors' content each AI summary includes, based on overlapping words. Which metric fits BEST?