Loading [MathJax]/extensions/tex2jax.js

Topic space trajectories: A case study on machine learning literature

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

  • Bastian Schaefermeier
  • Gerd Stumme
  • Tom Hanika

Organisationseinheiten

Externe Organisationen

  • Universität Kassel
Plum Print visual indicator of research metrics
  • Citations
    • Citation Indexes: 1
  • Captures
    • Readers: 16
see details

Details

OriginalspracheEnglisch
Seiten (von - bis)5759-5795
Seitenumfang37
FachzeitschriftSCIENTOMETRICS
Jahrgang126
Ausgabenummer7
Frühes Online-Datum15 Mai 2021
PublikationsstatusVeröffentlicht - Juli 2021

Abstract

The annual number of publications at scientific venues, for example, conferences and journals, is growing quickly. Hence, even for researchers it becomes harder and harder to keep track of research topics and their progress. In this task, researchers can be supported by automated publication analysis. Yet, many such methods result in uninterpretable, purely numerical representations. As an attempt to support human analysts, we present topic space trajectories, a structure that allows for the comprehensible tracking of research topics. We demonstrate how these trajectories can be interpreted based on eight different analysis approaches. To obtain comprehensible results, we employ non-negative matrix factorization as well as suitable visualization techniques. We show the applicability of our approach on a publication corpus spanning 50 years of machine learning research from 32 publication venues. In addition to a thorough introduction of our method, our focus is on an extensive analysis of the results we achieved. Our novel analysis method may be employed for paper classification, for the prediction of future research topics, and for the recommendation of fitting conferences and journals for submitting unpublished work. An advantage in these applications over previous methods lies in the good interpretability of the results obtained through our methods.

ASJC Scopus Sachgebiete

Zitieren

Topic space trajectories: A case study on machine learning literature. / Schaefermeier, Bastian; Stumme, Gerd; Hanika, Tom.
in: SCIENTOMETRICS, Jahrgang 126, Nr. 7, 07.2021, S. 5759-5795.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Schaefermeier B, Stumme G, Hanika T. Topic space trajectories: A case study on machine learning literature. SCIENTOMETRICS. 2021 Jul;126(7):5759-5795. Epub 2021 Mai 15. doi: 10.1007/s11192-021-03931-0
Schaefermeier, Bastian ; Stumme, Gerd ; Hanika, Tom. / Topic space trajectories : A case study on machine learning literature. in: SCIENTOMETRICS. 2021 ; Jahrgang 126, Nr. 7. S. 5759-5795.
Download
@article{ecf77d0be4aa494c8c46dcae6cf1a3b6,
title = "Topic space trajectories: A case study on machine learning literature",
abstract = "The annual number of publications at scientific venues, for example, conferences and journals, is growing quickly. Hence, even for researchers it becomes harder and harder to keep track of research topics and their progress. In this task, researchers can be supported by automated publication analysis. Yet, many such methods result in uninterpretable, purely numerical representations. As an attempt to support human analysts, we present topic space trajectories, a structure that allows for the comprehensible tracking of research topics. We demonstrate how these trajectories can be interpreted based on eight different analysis approaches. To obtain comprehensible results, we employ non-negative matrix factorization as well as suitable visualization techniques. We show the applicability of our approach on a publication corpus spanning 50 years of machine learning research from 32 publication venues. In addition to a thorough introduction of our method, our focus is on an extensive analysis of the results we achieved. Our novel analysis method may be employed for paper classification, for the prediction of future research topics, and for the recommendation of fitting conferences and journals for submitting unpublished work. An advantage in these applications over previous methods lies in the good interpretability of the results obtained through our methods.",
keywords = "Interpretable Machine Learning, Multidimensional Scaling, Non-Negative Matrix Factorization, Publication Dynamics, Topic Models",
author = "Bastian Schaefermeier and Gerd Stumme and Tom Hanika",
note = "Funding Information: This work was funded by the German Federal Ministry of Education and Research (BMBF) in its program “Quantitative Wissenschaftsforschung” as part of the REGIO project under Grant 01PU17012 as well as the German Research Foundation (DFG) priority programme (SPP) 1894 project topikos. Open Access funding enabled and organized by Projekt DEAL. ",
year = "2021",
month = jul,
doi = "10.1007/s11192-021-03931-0",
language = "English",
volume = "126",
pages = "5759--5795",
journal = "SCIENTOMETRICS",
issn = "0138-9130",
publisher = "Springer Netherlands",
number = "7",

}

Download

TY - JOUR

T1 - Topic space trajectories

T2 - A case study on machine learning literature

AU - Schaefermeier, Bastian

AU - Stumme, Gerd

AU - Hanika, Tom

N1 - Funding Information: This work was funded by the German Federal Ministry of Education and Research (BMBF) in its program “Quantitative Wissenschaftsforschung” as part of the REGIO project under Grant 01PU17012 as well as the German Research Foundation (DFG) priority programme (SPP) 1894 project topikos. Open Access funding enabled and organized by Projekt DEAL.

PY - 2021/7

Y1 - 2021/7

N2 - The annual number of publications at scientific venues, for example, conferences and journals, is growing quickly. Hence, even for researchers it becomes harder and harder to keep track of research topics and their progress. In this task, researchers can be supported by automated publication analysis. Yet, many such methods result in uninterpretable, purely numerical representations. As an attempt to support human analysts, we present topic space trajectories, a structure that allows for the comprehensible tracking of research topics. We demonstrate how these trajectories can be interpreted based on eight different analysis approaches. To obtain comprehensible results, we employ non-negative matrix factorization as well as suitable visualization techniques. We show the applicability of our approach on a publication corpus spanning 50 years of machine learning research from 32 publication venues. In addition to a thorough introduction of our method, our focus is on an extensive analysis of the results we achieved. Our novel analysis method may be employed for paper classification, for the prediction of future research topics, and for the recommendation of fitting conferences and journals for submitting unpublished work. An advantage in these applications over previous methods lies in the good interpretability of the results obtained through our methods.

AB - The annual number of publications at scientific venues, for example, conferences and journals, is growing quickly. Hence, even for researchers it becomes harder and harder to keep track of research topics and their progress. In this task, researchers can be supported by automated publication analysis. Yet, many such methods result in uninterpretable, purely numerical representations. As an attempt to support human analysts, we present topic space trajectories, a structure that allows for the comprehensible tracking of research topics. We demonstrate how these trajectories can be interpreted based on eight different analysis approaches. To obtain comprehensible results, we employ non-negative matrix factorization as well as suitable visualization techniques. We show the applicability of our approach on a publication corpus spanning 50 years of machine learning research from 32 publication venues. In addition to a thorough introduction of our method, our focus is on an extensive analysis of the results we achieved. Our novel analysis method may be employed for paper classification, for the prediction of future research topics, and for the recommendation of fitting conferences and journals for submitting unpublished work. An advantage in these applications over previous methods lies in the good interpretability of the results obtained through our methods.

KW - Interpretable Machine Learning

KW - Multidimensional Scaling

KW - Non-Negative Matrix Factorization

KW - Publication Dynamics

KW - Topic Models

UR - http://www.scopus.com/inward/record.url?scp=85106227550&partnerID=8YFLogxK

U2 - 10.1007/s11192-021-03931-0

DO - 10.1007/s11192-021-03931-0

M3 - Article

AN - SCOPUS:85106227550

VL - 126

SP - 5759

EP - 5795

JO - SCIENTOMETRICS

JF - SCIENTOMETRICS

SN - 0138-9130

IS - 7

ER -