Applications of machine learning in cannabis research: A scoping review
Bibliographic record
Abstract
• Machine learning (ML) is increasingly applied in cannabis research for data analysis. • ML algorithms predict risks and factors contributing to cannabis use effectively. • ML tools enable effective monitoring and information extraction about cannabis. • Bias arises from cross-sectional, non-representative data and recall-based training. • Re -evaluating methods and validating ML models may improve research applicability. Over the past decade, research about cannabis and its associated compounds has increased substantially. Machine learning (ML) is increasingly used in cannabis-related research to improve data analysis and modeling. The present scoping review aimed to identify how ML is used in the context of cannabis research. A scoping review was conducted following Arksey and O'Malley's five-stage scoping review framework. MEDLINE, EMBASE, PsycINFO and CINAHL were systematically searched, and CADTH was searched using keywords. Studies utilizing ML in the context of cannabis research were deemed eligible. Title and abstract and full text screening, data extraction, thematic coding, and analysis were performed independently and in duplicate for all included studies. Forty-six studies were included. Four themes emerged: 1) the sampling methodologies utilized in studies investigating cannabis and ML introduce bias in results, 2) ML algorithms can predict characteristics associated with cannabis use, including predictive factors, risk of usage, and impact on users, 3) ML algorithms are an effective tool for monitoring and extracting information about cannabis; and 4) various ML algorithms were most suitable for different tasks. This scoping review highlights two major uses of ML algorithms in cannabis research—for predicting risks of and factors contributing to cannabis use, and for extracting information about cannabis. Challenges associated with ML in cannabis research included the introduction of bias in results from the use of cross-sectional and non-representative data, and recall bias which may have led to biased training of ML models. Re -evaluating study methodology suitability and externally validating ML models may increase the viability/applicability of ML in cannabis research.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.023 | 0.106 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.005 |
| Bibliometrics | 0.029 | 0.030 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.006 | 0.006 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.004 | 0.003 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".