Improving author coreference by resource-bounded information gathering from the web

Abstract

Accurate entity resolution is sometimes impossi-ble simply due to insufficient information. For ex-ample, in research paper author name resolution, even clever use of venue, title and co-authorship relations are often not enough to make a confi-dent coreference decision. This paper presents sev-eral methods for increasing accuracy by gather-ing and integrating additional evidence from the web. We formulate the coreference problem as one of graph partitioning with discriminatively-trained edge weights, and then incorporate web informa-tion either as additional features or as additional nodes in the graph. Since the web is too large to incorporate all its data, we need an efficient pro-cedure for selecting a subset of web queries and data. We formally describe the problem of re-source bounded information gathering in each of these contexts, and show significant accuracy im-provement with low cost. 1

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 66 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.