2

Example: the top conferences/journals for computer vision have an h5-index almost twice higher than the top conferences/journals for natural language processing:

https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_computervisionpatternrecognition (mirror 1, mirror 2):

Index Publication h5-index h5-median
1. IEEE/CVF Conference on Computer Vision and Pattern Recognition 299 509
2. IEEE/CVF International Conference on Computer Vision 176 295
3. European Conference on Computer Vision 144 286
4. IEEE Transactions on Pattern Analysis and Machine Intelligence 131 261
5. IEEE Transactions on Image Processing 113 156
6. Pattern Recognition 85 126
7. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops 73 110
8. International Journal of Computer Vision 70 150
9. Medical Image Analysis 67 115
10. Pattern Recognition Letters 59 80

https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_computationallinguistics (mirror 1, mirror 2):

# Publication h5-index h5-median
1. Meeting of the Association for Computational Linguistics (ACL) 135 220
2. Conference on Empirical Methods in Natural Language Processing (EMNLP) 112 197
3. Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (HLT-NAACL) 90 148
4. Transactions of the Association for Computational Linguistics 53 112
5. International Workshop on Semantic Evaluation 50 93
6. International Conference on Computational Linguistics (COLING) 49 73
7. Conference of the European Chapter of the Association for Computational Linguistics (EACL) 45 68
8. Conference on Computational Natural Language Learning (CoNLL) 39 68
9. International Conference on Language Resources and Evaluation (LREC) 38 54
10. Workshop on Machine Translation 35 64
Franck Dernoncourt
  • 33,669
  • 27
  • 144
  • 313

1 Answers1

9

Different fields have different sizes, speed and publication culture. Thus, it is completely expected that h5-indices will vary a lot between fields. This effect is going to be more pronounced when larger values are considered.

Size: To push an h5-index from eg 199 to 200, the additional paper needs to be cited 200 times within 5 years. In a smaller field, this could easily mean getting citations from an absurdly high fraction of the new papers published.

Speed: The h5-index only counts citations in a 5-year window. In a fast moving field, that doesn't matter much. In a slower-moving field, this can easily ignore most of the potential citations a paper could get.

Publication culture: How big or small are the units people publish? How many references does a typical paper have? Do authors always explicitly cite the masters, or only if really needed?

Arno
  • 43,962
  • 8
  • 123
  • 169