Anchor Prediction: Automatic Refinement of Internet Links
This addresses the issue for internet users who waste effort finding relevant information in linked webpages, but it is incremental as it builds on existing NLP and ranking methods.
The paper tackles the problem of unanchored internet links by introducing the task of anchor prediction to identify specific parts of linked webpages related to the source context, and it releases two datasets (AuthorAnchors with 34K links and ReaderAnchors) while benchmarking a T5-based approach that shows room for improvement.
Internet links enable users to deepen their understanding of a topic by providing convenient access to related information. However, the majority of links are unanchored -- they link to a target webpage as a whole, and readers may expend considerable effort localizing the specific parts of the target webpage that enrich their understanding of the link's source context. To help readers effectively find information in linked webpages, we introduce the task of anchor prediction, where the goal is to identify the specific part of the linked target webpage that is most related to the source linking context. We release the AuthorAnchors dataset, a collection of 34K naturally-occurring anchored links, which reflect relevance judgments by the authors of the source article. To model reader relevance judgments, we annotate and release ReaderAnchors, an evaluation set of anchors that readers find useful. Our analysis shows that effective anchor prediction often requires jointly reasoning over lengthy source and target webpages to determine their implicit relations and identify parts of the target webpage that are related but not redundant. We benchmark a performant T5-based ranking approach to establish baseline performance on the task, finding ample room for improvement.