Environmental-Health Convergence: A deep learning-oriented decision support system for catalyzing sustainable healthy food systems
Bibliographic record
Abstract
To generate evidence to address food system challenges, we developed an adaptable framework for multimodel assessment of the convergence effect of health and environmental drivers in food systems. We achieved this goal by developing a modeling framework that facilitates testing and applying four deep-learning algorithms using a case study of the United States's food system. Among the models tested, the bidirectional and single-layer long short-term memory models outperformed the others with α E (2.75) and α H (3.51) when predicting environmental drivers and health drivers, respectively. All the models tested performed better at predicting environmental than health drivers. The best-performing model for each dimension was deployed into the Food System Rapid Overview Assessment through Scenarios (FS-ROAS) tool. As we approach the endpoint of the transformative 2030 agenda, FS-ROAS can be a timely toolkit that enables stakeholders to explore diverse intervention scenarios in the context of short-medium and long-term goals for future food systems and generate evidence to guide future actions. • We built a framework to test and apply deep learning models for multioutput predictions. • The LSTM model outperformed all other models when predicting health indicators. • The BiLSTM model outperformed all other models when predicting environmental indicators. • The LSTM and BiLSTM models were deployed into a decision support system (DSS). • The DSS generates evidence to guide public health and climate mitigation strategies.
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 | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| 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".