Enhancing Relational Topic Models with Named Entity Induced Links.

Felix Kuhr; Mathis Lichtenberger; Tanya Braun; Ralf Möller

doi:10.1109/ICSC50631.2021.00059

Enhancing Relational Topic Models with Named Entity Induced Links.

Felix Kuhr, Mathis Lichtenberger, Tanya Braun, Ralf Möller

Abstract

Relational topic modeling as an extension to classical topic modeling assumes that documents with some form of link between the documents share topics. The links between documents are given from hyperlinks in web documents, citations in articles, or friendships in social networks. In this work, we consider links between documents induced from named entities: Two documents are linked to each other if both documents have a named entity in common. We present a case study on the performance of relational topic modeling using named-entity induced links between documents. Comparing the prediction accuracy with different sets of named-entity induced links, the results show that additional links between documents can increase the performance of topic models.

Original language	English
Title of host publication	ICSC
Number of pages	4
Publication date	2021
Pages	314-317
DOIs	https://doi.org/10.1109/ICSC50631.2021.00059
Publication status	Published - 2021

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 4 Quality Education
SDG 9 Industry, Innovation, and Infrastructure
SDG 11 Sustainable Cities and Communities
SDG 12 Responsible Consumption and Production
SDG 14 Life Below Water
SDG 15 Life on Land

Research Areas and Centers

Centers: Center for Artificial Intelligence Luebeck (ZKIL)
Research Area: Intelligent Systems

Access to Document

10.1109/ICSC50631.2021.00059

Cite this

@inproceedings{84d867130b3b4faa9805bd7285b9cab1,

title = "Enhancing Relational Topic Models with Named Entity Induced Links.",

abstract = "Relational topic modeling as an extension to classical topic modeling assumes that documents with some form of link between the documents share topics. The links between documents are given from hyperlinks in web documents, citations in articles, or friendships in social networks. In this work, we consider links between documents induced from named entities: Two documents are linked to each other if both documents have a named entity in common. We present a case study on the performance of relational topic modeling using named-entity induced links between documents. Comparing the prediction accuracy with different sets of named-entity induced links, the results show that additional links between documents can increase the performance of topic models.",

author = "Felix Kuhr and Mathis Lichtenberger and Tanya Braun and Ralf M{\"o}ller",

note = "DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.",

year = "2021",

doi = "10.1109/ICSC50631.2021.00059",

language = "English",

pages = "314--317",

booktitle = "ICSC",

}

TY - GEN

T1 - Enhancing Relational Topic Models with Named Entity Induced Links.

AU - Kuhr, Felix

AU - Lichtenberger, Mathis

AU - Braun, Tanya

AU - Möller, Ralf

N1 - DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.

PY - 2021

Y1 - 2021

N2 - Relational topic modeling as an extension to classical topic modeling assumes that documents with some form of link between the documents share topics. The links between documents are given from hyperlinks in web documents, citations in articles, or friendships in social networks. In this work, we consider links between documents induced from named entities: Two documents are linked to each other if both documents have a named entity in common. We present a case study on the performance of relational topic modeling using named-entity induced links between documents. Comparing the prediction accuracy with different sets of named-entity induced links, the results show that additional links between documents can increase the performance of topic models.

AB - Relational topic modeling as an extension to classical topic modeling assumes that documents with some form of link between the documents share topics. The links between documents are given from hyperlinks in web documents, citations in articles, or friendships in social networks. In this work, we consider links between documents induced from named entities: Two documents are linked to each other if both documents have a named entity in common. We present a case study on the performance of relational topic modeling using named-entity induced links between documents. Comparing the prediction accuracy with different sets of named-entity induced links, the results show that additional links between documents can increase the performance of topic models.

U2 - 10.1109/ICSC50631.2021.00059

DO - 10.1109/ICSC50631.2021.00059

M3 - Conference contribution

SP - 314

EP - 317

BT - ICSC

ER -

Enhancing Relational Topic Models with Named Entity Induced Links.

Abstract

UN SDGs

Research Areas and Centers

Access to Document

Fingerprint

Cite this