| 1. |
- Ahlgren, Per, et al.
(författare)
-
Community Detection Using Citation Relations and Textual Similarities in a Large Set of PubMed Publications
- 2019
-
Ingår i: PROCEEDINGS OF THE 17th INTERNATIONAL CONFERENCE ON SCIENTOMETRICS & INFORMETRICS. - Leuven : International Society for Scientometrics and Informetrics. - 9788833811185 ; , s. 1380-1392
-
Konferensbidrag (refereegranskat)abstract
- In this contribution, the effects of enhancing direct citations, with respect to publication-publication relatedness measurement, by indirect citation relations (bibliographic coupling and co-citation) and text relations on clustering accuracy are analyzed. In total, we investigate six approaches. In one of these, direct citations are enhanced by both bibliographic coupling and co-citation, whereas text relations are used to enhance direct citations in another approach. In addition to an approach based on direct citations only, we include in the study, for comparison reasons, each approach that is involved in the enhancement of direct citations. For the evaluation of the approaches, we use a methodology proposed by earlier research. However, the used evaluation criterion is based on MeSH, arguable the most sophisticated item-level classification scheme available. The results show that the co-citation approach has the worst performance, and that the direct citations approach is outperformed by the other four investigated approaches. An approach in which direct citations are enhanced by the BM25 textual relatedness measure has the best performance, followed by the approach that combines direct citations with bibliographic coupling and co-citation. The latter performs slightly better than the bibliographic coupling approach, which in turn has a better performance than the BM25 approach.
|
|
| 2. |
|
|
| 3. |
- Eklund, Johan, et al.
(författare)
-
MESH classification of clinical guidelinesusing conceptual embeddings of references
- 2019
-
Ingår i: Proceedings of the 17th conference of the International society for scientometrics and informetrics, ISSI. - 9788833811185 ; , s. 859-864
-
Konferensbidrag (refereegranskat)abstract
- In this study, we investigate different strategies for assigning MeSH (Medical Subject Headings) terms to clinical guidelines using machine learning. Features based on words in titles and abstracts are investigated and compared to features based on topics assigned to references cited by the guidelines. Two of the feature engineering strategies utilize word embeddings produced by recent models based on in the distributional hypothesis, called word2vecand fastText. The evaluation results show that reference-based strategies tend to yield a higher recall and F1 scores for MeSH terms with a sufficient amount of training instances, whereas title and abstract based features yield a higher precision.
|
|
| 4. |
- Jarneving, Bo, 1956
(författare)
-
Elaborations on a cluster analytical approach to author bibliographic coupling analysis in the context of science mapping
- 2019
-
Ingår i: 17th International Conference of the International-Society-for-Scientometrics-and-Informetrics (ISSI) on Scientometrics and Informetrics (Vol. 1). September 02-05, 2019. Rome, Italy. - : ISSI. - 2175-1935. - 9788833811185
-
Konferensbidrag (refereegranskat)abstract
- Previous research on bibliographic coupling has focused on the research article as the analysed unit. In later research the author has been put forward as the analysed unit and the term author bibliographic coupling analysis (ABCA) has been coined. in the context of science mapping, factor analysis and network analysis have been applied in combination with various methods of counting author bibliographic coupling units. These method approaches are contrasted with a cluster analytical one, applying two recognized similarity measures, the cosine similarity measure and the Jaccard index, in combination with two different cluster methods, the SLM-algorithm and Ward's method. The purpose of this approach was to test the suitability of ABCA as a tool for deriving classifications in the context of science mapping. Findings revealed that the cosine similarity measure provided with the most valid cluster solution for both cluster methods. The optimal method combination comprised Ward's method and the cosine similarity measure. However, the evaluation of clusters' quality disclosed a lack of structure in data, suggesting that cluster analysis is not a suitable method in the context of ABCA. Considering the features of author bibliographic coupling data, the suitability of ABCA in the context of science mapping was challenged.
|
|
| 5. |
|
|
| 6. |
- Rahman, Jakaria, 1977
(författare)
-
Does collaborative research published in top journals remain uncited?
- 2019
-
Ingår i: 17th International Conference on Scientometrics and Informetrics, ISSI 2019 - Proceedings. - 9788833811185 ; 1, s. 1068-1074
-
Konferensbidrag (refereegranskat)abstract
- This paper investigates whether collaborative research published in top journals remains uncited, and to what extent access type (open and closed) affects on citation of collaborative research published in top journals. It looks at publications including articles, conference papers, reviews, short surveys, editorials, letters, notes published between 2009-2016 with an affiliation from Chalmers University of Technology and indexed in Scopus. To give enough time to gather citation, two-year time frame is considered for the publication of the year 2016. The data is classified based on access types: closed and open access, and sub-classified as cited closed access, cited open access, non-cited closed access, and non-cited open access in SciVal. The top 25 percentile indicating the number of journals that are in the top 25% of the most cited journals indexed by Scopus is considered. The result shows that a small portion of collaborative research published in top journals remain uncited irrespective of types of collaboration. In case of international collaborative research, publications in closed access are more cited than in open access. Institutional collaborative research publications are more cited than national collaborative ones. Collaborative research is more cited than single authors’ publications and single authored conference papers published in the top percentile do not remain uncited.
|
|
| 7. |
- Sandström, Ulf, 1953-, et al.
(författare)
-
Performance of Research Teams : results from 107 European groups
- 2019
-
Ingår i: 17TH INTERNATIONAL CONFERENCE ON SCIENTOMETRICS & INFORMETRICS (ISSI2019), VOL II. - : INT SOC SCIENTOMETRICS & INFORMETRICS-ISSI. - 9788833811185 ; , s. 2240-2251
-
Konferensbidrag (refereegranskat)abstract
- This paper investigates what factors affect the performance of research teams. We combine survey data about the team with bibliometric data about the performance of the team. The analysis shows that teams with a few PIs perform better than single PI teams - of course controlling for team size. On the other hand, gender diversity does not have an effect on performance. The good news is that gender objectives can be realized, without any performance problem.
|
|
| 8. |
- van den Besselaar, Peter, et al.
(författare)
-
Recognition through performance and reputation
- 2019
-
Ingår i: 17TH INTERNATIONAL CONFERENCE ON SCIENTOMETRICS & INFORMETRICS (ISSI2019), VOL II. - : INT SOC SCIENTOMETRICS & INFORMETRICS-ISSI. - 9788833811185 ; , s. 2065-2069
-
Konferensbidrag (refereegranskat)abstract
- As the various disciplines have different forms of social and intellectual organization (Whitley 2000), scholars in various fields may depend less on their peers, and more on other audiences for recognition and funding. Following Merton (1973) we distinguish between performance and reputation for building up recognition. We show that there are indeed differences between the disciplines: in life sciences and social sciences, the reputation related indicators are dominant in predicting the score that grant applicants get from the panel, whereas in the natural sciences, the performance related indicators dominate the panel scores. Furthermore, when comparing within the life sciences the grantees with the best performing non-grantees, we show that the former score higher on the reputation indicators and the second score better on the performance variables, supporting the findings that in life sciences one probably gains recognition over reputation more than over individual performance. We suggest that this may not be optimal for the growth of knowledge.
|
|
| 9. |
- Åström, Fredrik, et al.
(författare)
-
Conceptualising dimensions of bibliometric assessment: From resource allocation systems to evaluative landscapes
- 2019
-
Ingår i: Proceedings of the 17th Conference of the International Society for Scientometrics and Informetrics : ISSI 2019 with a Special STI Indicators Conference Track - ISSI 2019 with a Special STI Indicators Conference Track. - 9788833811185 ; , s. 1256-1261
-
Konferensbidrag (refereegranskat)abstract
- The purpose of this paper is to discuss the conceptualisation of bibliometric analyses in terms of the levels on which they are performed, adding contextual factors to the dimension where the size of the unit being analysed is considered. Based on empirical investigations of resource allocation systems and research evaluation practices, as well as the previous literature conceptualising bibliometric analyses, a framework based on Whitley’s (2000) notion of research fields as ‘reputational work organisations’, is discussed. The results suggest adding a contextual ‘reputational dimension’ to the size-based dimension distinguishing between micro-, meso- and macro-level analyses. Furthermore, we propose that ‘evaluative landscapes’ (Brandter, 2017) might be a fruitful approach for further analysing how complex and multifaceted landscapes of research assessment affects the individual researcher.
|
|
