Global Machine-learning Research: a scientometric assessment of global literature during 2009–18
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The article provides a quantitative and qualitative analyses of global machine-learning research output (48,455 publications), using select bibliometric indicators, using Web of Science database for 2009–18 period. The various indicators used in this study are: average annual growth, citations per paper, international collaborative papers, relative citation index, activity index, top-productive countries, organizations, authors, journals, and highly cited papers. Machine learning (within the domain of artificial intelligence) as a subject of study has fast-emerged as a subject of intensive research. It registered average annual growth rate of 27.59% and averaged citation impact of 10.78 per paper. Among 138 participating countries, the USA and China were in top 10 most productive countries on the subject. Among top 10 countries, France and Canada were the leading countries in terms of average citation per paper and relative index. France and Australia were leading in terms of for their national-level share to international collaborative publications (64.95% and 63.95%, respectively). In terms of type of machine learning, supervised learning registered the largest publications’ share, followed by deep learning, semi-supervised learning and reinforced learning (0.89% share, 556 papers). Centre National De La Recherche Scientique, France (769 papers), Harvard University, USA (751 papers) and University of London, UK (729 papers) were the three most productive global research organizations. In contrast, University of Toronto, Canada, Nanyang Technological University, Singapore and University of Oxford, UK were the three leading organizations in terms of citation per paper and relative citation index. Y. Zhang (246 papers), Y. Liu (204 papers) and J. Wang (203 papers) were leading in publication productivity in contrast to J. Li (12.52 and 1.03). L. Zhang (12.42 and 1.02) and J. Zhang (11.23 and 0.92) scored high in citation per paper and relative citation index on the subject. Neurocomputing (1310 papers), PLOS One (917 papers) and Expert Systems with Applications (861 papers) were the leading journals on this subject.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Direct model labels (unvalidated)
Per-model category and study-design labels from the labeling rounds. They are machine output, unvalidated, and the disagreement between models ships as data. No study design here is MEDLINE-validated yet.
| Model arm | Categories | Study design | Confidence |
|---|---|---|---|
| gemma | Bibliometrics Domain: not available · Genre: Empirical About the Canadian research system: no · About a Canadian topic: no | Observational | low |
| gpt | Bibliometrics Domain: not available · Genre: Empirical About the Canadian research system: no · About a Canadian topic: no | Observational | high |
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it