File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1145/2983323.2983730
- Scopus: eid_2-s2.0-84996527452
- WOS: WOS:000390890800085
Supplementary
- Citations:
- Appears in Collections:
Conference Paper: On transductive classification in heterogeneous information networks
Title | On transductive classification in heterogeneous information networks |
---|---|
Authors | |
Keywords | Heterogeneous information network Transductive classification Knowledge base |
Issue Date | 2016 |
Publisher | ACM. |
Citation | The 25th ACM International Conference on Information and Knowledge Management (CIKM 2016), Indianapolis, IN., 24-28 October 2016. In Conference Proceedings, 2016, p. 1-10 How to Cite? |
Abstract | A heterogeneous information network (HIN) is used to model objects of different types and their relationships. Objects are often associated with properties such as labels. In many applications, such as curated knowledge bases for which object labels are manually given, only a small fraction of the objects are labeled. Studies have shown that transductive classification is an effective way to classify and to deduce labels of objects, and a number of transductive classifiers have been put forward to classify objects in an HIN. We study the performance of a few representative transductive classification algorithms on HINs. We identify two fundamental properties, namely, cohesiveness and connectedness, of an HIN that greatly influence the effectiveness of transductive classifiers. We define metrics that measure the two properties. Through experiments, we show that the two properties serve as very effective indicators that predict the accuracy of transductive classifiers. Based on cohesiveness and connectedness we derive (1) a black-box tester that evaluates whether transductive classifiers should be applied for a given classification task and (2) an active learning algorithm that identifies the objects in an HIN whose labels should be sought in order to improve classification accuracy. |
Persistent Identifier | http://hdl.handle.net/10722/232180 |
ISBN | |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Li, X | - |
dc.contributor.author | Kao, BCM | - |
dc.contributor.author | Zheng, Y | - |
dc.contributor.author | Huang, Z | - |
dc.date.accessioned | 2016-09-20T05:28:17Z | - |
dc.date.available | 2016-09-20T05:28:17Z | - |
dc.date.issued | 2016 | - |
dc.identifier.citation | The 25th ACM International Conference on Information and Knowledge Management (CIKM 2016), Indianapolis, IN., 24-28 October 2016. In Conference Proceedings, 2016, p. 1-10 | - |
dc.identifier.isbn | 978-1-4503-4073 | - |
dc.identifier.uri | http://hdl.handle.net/10722/232180 | - |
dc.description.abstract | A heterogeneous information network (HIN) is used to model objects of different types and their relationships. Objects are often associated with properties such as labels. In many applications, such as curated knowledge bases for which object labels are manually given, only a small fraction of the objects are labeled. Studies have shown that transductive classification is an effective way to classify and to deduce labels of objects, and a number of transductive classifiers have been put forward to classify objects in an HIN. We study the performance of a few representative transductive classification algorithms on HINs. We identify two fundamental properties, namely, cohesiveness and connectedness, of an HIN that greatly influence the effectiveness of transductive classifiers. We define metrics that measure the two properties. Through experiments, we show that the two properties serve as very effective indicators that predict the accuracy of transductive classifiers. Based on cohesiveness and connectedness we derive (1) a black-box tester that evaluates whether transductive classifiers should be applied for a given classification task and (2) an active learning algorithm that identifies the objects in an HIN whose labels should be sought in order to improve classification accuracy. | - |
dc.language | eng | - |
dc.publisher | ACM. | - |
dc.relation.ispartof | ACM International Conference on Information and Knowledge Management, CIKM'16 Proceedings | - |
dc.subject | Heterogeneous information network | - |
dc.subject | Transductive classification | - |
dc.subject | Knowledge base | - |
dc.title | On transductive classification in heterogeneous information networks | - |
dc.type | Conference_Paper | - |
dc.identifier.email | Kao, BCM: kao@cs.hku.hk | - |
dc.identifier.authority | Kao, BCM=rp00123 | - |
dc.identifier.doi | 10.1145/2983323.2983730 | - |
dc.identifier.scopus | eid_2-s2.0-84996527452 | - |
dc.identifier.hkuros | 264588 | - |
dc.identifier.spage | 1 | - |
dc.identifier.epage | 10 | - |
dc.identifier.isi | WOS:000390890800085 | - |
dc.publisher.place | United States | - |
dc.customcontrol.immutable | sml 161017 | - |