Topic space trajectories
| dc.date.accessioned | 2021-07-21T12:49:06Z | |
| dc.date.available | 2021-07-21T12:49:06Z | |
| dc.date.issued | 2021-05-15 | |
| dc.description.sponsorship | Gefördert im Rahmen des Projekts DEAL | ger |
| dc.identifier | doi:10.17170/kobra-202107054243 | |
| dc.identifier.uri | http://hdl.handle.net/123456789/13030 | |
| dc.language.iso | eng | eng |
| dc.relation.doi | doi:10.1007/s11192-021-03931-0 | |
| dc.rights | Namensnennung 4.0 International | * |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | * |
| dc.subject | topic models | eng |
| dc.subject | non-negative matrix factorization | eng |
| dc.subject | multidimensional scaling | eng |
| dc.subject | publication dynamics | eng |
| dc.subject | interpretable machine learning | eng |
| dc.subject.ddc | 004 | |
| dc.subject.swd | Thema | ger |
| dc.subject.swd | Wissenschaftliche Literatur | ger |
| dc.subject.swd | Literaturproduktion | ger |
| dc.subject.swd | Maschinelles Lernen | ger |
| dc.subject.swd | Multidimensionale Skalierung | ger |
| dc.subject.swd | Nichtnegative Matrix | ger |
| dc.title | Topic space trajectories | eng |
| dc.title.subtitle | A case study on machine learning literature | eng |
| dc.type | Aufsatz | |
| dc.type.version | publishedVersion | |
| dcterms.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. | eng |
| dcterms.accessRights | open access | |
| dcterms.creator | Schäfermeier, Bastian | |
| dcterms.creator | Stumme, Gerd | |
| dcterms.creator | Hanika, Tom | |
| dcterms.source.identifier | eissn:1588-2861 | |
| dcterms.source.issue | Issue 7 | |
| dcterms.source.journal | Scientometrics | eng |
| dcterms.source.pageinfo | 5759-5795 | |
| dcterms.source.volume | Volume 126 | |
| kup.iskup | false |
Files
Original bundle
1 - 1 of 1
- Name:
- Schaefermeier2021_Article_TopicSpaceTrajectories.pdf
- Size:
- 3.85 MB
- Format:
- Adobe Portable Document Format
- Description:
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 3.03 KB
- Format:
- Item-specific license agreed upon to submission
- Description: