Change-Oriented Summarization of Temporal Scholarly Document Collections by Semantic Evolution Analysis

The number of scholarly publications has dramatically increased over the last decades. For anyone new to a particular science domain it is not easy understand major trends and significant changes that undergone time. Temporal summarization related approaches should be then useful make sense temporal collections. In this paper we demonstrate an approach analyze dataset research papers by providing high level overview important occurred time in dataset. novelty our lies adaptation methods used for semantic term evolution analysis. However, just single words independently, but estimate common drifts shared groups semantically converging words. As example study ACL Anthology Reference Corpus spans from 1974 2015 contains 22,878 articles.