An Efficient TabNet-Based Obesity Classification and Risk Prediction System Optimized with Dove Flocking Optimizer
Bibliographic record
Abstract
Obesity is on the rise, and it has turned out to be a big challenge in healthcare. Simple and straightforward models are necessary to diagnose it. Thus, designed an arrangement named EfficientTabNet in the Obesity Diagnosis and Decision System Optimised with Dove Flocking Optimizer (ETN-OD-DFO). To clean up the data using the High Accuracy Distributed Kalman Filter (HADKF) first, though. It assists in eliminating noise, equalising the data, and ensuring it is all consistent. The processed data is analysed using the Efficient TabNet (ETN) model. This is an attention-based model that can be used to select significant features and easily integrate multiple kinds of data. The settings are also adjusted with the Dove Flocking Optimizer (DFO). It is a bio inspired approach which assists in the search of the optimal solutions by imitating the process of flocking of birds. Our system has been tested on the Obesity Dataset having 2,111 records and 17 features. Our model achieves approximately 97.46-percent accuracy, 97.12-percent precision and 97.05-percent F1-score. It wins over the fundamental ETN by approximately$6-8 \%$. Also, the MCC and Kappa scores increased by more than 9 percent and the log loss decreased by 82.22. This demonstrates the effectiveness of using ETN-OD-DFO system in diagnosing obesity. Biology is easy to comprehend and inspired. It is a full 4 seconds slower to execute, although it usually produces more impressive results.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 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".