When summarizing a collection of views, arguments or opinions on some topic,\nit is often desirable not only to extract the most salient points, but also to\nquantify their prevalence. Work on multi-document summarization has\ntraditionally focused on creating textual summaries, which lack this\nquantitative aspect. Recent work has proposed to summarize arguments by mapping\nthem to a small set of expert-generated key points, where the salience of each\nkey point corresponds to the number of its matching arguments. The current work\nadvances key point analysis in two important respects: first, we develop a\nmethod for automatic extraction of key points, which enables fully automatic\nanalysis, and is shown to achieve performance comparable to a human expert.\nSecond, we demonstrate that the applicability of key point analysis goes well\nbeyond argumentation data. Using models trained on publicly available\nargumentation datasets, we achieve promising results in two additional domains:\nmunicipal surveys and user reviews. An additional contribution is an in-depth\nevaluation of argument-to-key point matching models, where we substantially\noutperform previous results.\n