Link Discovery: A Comprehensive Analysis

被引：5

作者：

Erbs, Nicolai ^{[1
]}

Zesch, Torsten ^{[1
]}

Gurevych, Iryna ^{[1
]}

机构：

[1] Tech Univ Darmstadt, Ubiquitous Knowledge Proc Lab, Darmstadt, Germany

来源：

FIFTH IEEE INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC 2011) | 2011年

关键词：

D O I：

10.1109/ICSC.2011.63

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a comprehensive analysis of link discovery approaches. We classify them with regard to the type of knowledge being used, and identify three commonly used sources of knowledge: The text of a document, the document title, and already existing links. We analyze the influence of the knowledge source as well as of the amount of training data used. Results show that the link-based approach performs best if the amount of training data is huge. In a more realistic setting with fewer training data, the text-based approach yields better results.

引用

页码：83 / 86

页数：4

共 10 条

[1] [Anonymous], 2007, Proceedings of the 16th ACM Conference on Con- ference on Information and Knowledge Management, DOI DOI 10.1145/1321440.1321475.19
[2] Buffa M., 2006, P INTRAWEBS WORKSH 2
[3] Geva S., 2007, PREPR INEX WORKSH, P404
[4] Hoffart Johannes, 2009, PREPR INEX WORKSH, P314
[5] Huang W.C., 2009, LECT NOTES COMPUTER, P312
[6] Itakura K.Y., 2007, INEX 2007 WORKSH PRE, P417
[7] Majchrzak A., 2006, P 2006 INT S WIKIS, P99
[8] Manning C., 1999, Foundations of Statistical Natural Language Processing
[9] Mihalcea R., 2004, P EMNLP 2004 ASS COM, P404
[10] Salton G., 1983, INTRO MODERN INFORM

← 1 →