HITS' Graph-based System at the NTCIR-9 Cross-lingual Link Discovery Task

Abstract

This paper presents HITS’ system for the NTCIR-9 crosslingual link discovery task. We solve the task in three stages: (1) anchor identification and ambiguity reduction, (2) graphbased disambiguation combining different relatedness measures as edge weights for a maximum edge weighted clique algorithm, and (3) supervised relevance ranking. In the fileto-file evaluation with Wikipedia ground-truth the HITS system is the top-performer across all measures and subtasks (English-2-Chinese, English-2-Japanese and English-2Korean). In the file-2-file and anchor-2-file evaluation with manual assessment, the system outperforms all other systems on the English-2-Japanese subtask and is one of the top-three performing systems for the two other subtasks.