J10-1002 differences of note in comparison to name disambiguation . First , in document classification
J10-1002 position of the target verb . In name disambiguation , we loosen the restriction to
J10-1002 under consideration . In proper name disambiguation ( Section 5 ) , we classify the
D12-1076 Wikipedia to enhance the results of name disambiguation . ( 4 ) SSR : the Structural
J10-1002 training approaches for SVM in name disambiguation . It is possible that the mapping
D12-1076 - tion , topic generation and name disambiguation . For an ambiguous name , we
J10-1002 gold-standard profiles . As in name disambiguation , we experiment with different
J10-1002 names thus serves as one of six name disambiguation tasks . Table 4 shows the number
D12-1076 time . Investigating the person name disambiguation task in different web applications
D09-1056 mention of a per son . In Web person name disambiguation it suffices to group the documents
J10-1002 distance to a target element ( as in name disambiguation ) . In each case , the resulting
J10-1002 classification task similarly to name disambiguation , taking a minimally supervised
J10-1002 gold-standard profiles only . For name disambiguation , we have 60 profiles ( 5 samplings
J10-1002 alternation detection ( Section 4 ) , name disambiguation ( Section 5 ) , and document
J10-1002 training and test sets . As in the name disambiguation task , we use the LIBSVM software
J10-1002 Interest in the NLP problem of name disambiguation has increased as the growth of
J10-1002 largest densities , followed by the name disambiguation data set , then the document
J10-1002 verb alternation detection and name disambiguation , but does poorly on document
J10-1002 perform well ( verb alternation and name disambiguation ) , and a low average value across
D12-1076 representations in the web person name disambiguation task which has been suffering
hide detail