id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_aptojhcwune4dnb3zom6igwm7e	Sourav Dutta	Cross-Document Co-Reference Resolution using Sample-Based Clustering with Knowledge Enrichment	2015	14	.pdf	application/pdf	9254	859	67	Cross-Document Co-Reference Resolution using Sample-Based Clustering problem in this context is cross-document coreference resolution (CCR): computing equivalence classes of textual mentions denoting It takes as input a set of documents with entity mentions, and computes as output a set of equivalence Cluster-ranking and multi-sieve methods incrementally expand groups of mentions and exploit Similarity computations between mention groups are performed lazily on-demand for the dynamically selected samples. • CROCS, a framework for cross-document coreference resolution using sample-based spectral mention groups ({mij}) obtained in the previous step, we combine the sentences of the mentions to determine the best matching entity in a (using the similarity metric) in a hierarchical fashion to compute the cross-document coreference equivalence classes of mentions. freebase.com entries, for possibly matching entities, to enrich the features of a mention group. It performs a topdown hierarchical bisection process, based on similarity scores among entities, to cluster together coreferring mention groups at each splitting level.	./cache/work_aptojhcwune4dnb3zom6igwm7e.pdf	./txt/work_aptojhcwune4dnb3zom6igwm7e.txt