Exploring Deep Learning–Based Models for Sociocultural African Food Recognition System
Bibliographic record
Abstract
Food recognition, a field under food computing, has significantly promoted people’s dietary decision‐making and culinary customs. We present the design and evaluation of a sociocultural app for African food recognition using deep learning models such as transfer learning. Deep learning models have multiple processing layers that make them robust in image recognition. Based on this capability of deep learning models, we explored them in this study. A total of 3142 food image datasets were collected from three African countries: Nigeria, Ghana, and Cameroon. Using the datasets, we developed and trained a deep learning model for recognizing African foods. The model attained a test accuracy of 94.5%. The model was further deployed in a food recognition app. To evaluate the predictive ability of the app, we recruited 16 participants who were interviewed and subsequently used the app in the wild for 7 days. In a comparative evaluation between the app and human recognition capabilities, we found that the app recognized 71% of the instances of food images generated by the participants and tested with the app, while the human evaluators (participants) could only recognize 56% of the food datasets. Participants were mostly able to recognize some foods from their own country. Furthermore, participants suggested some design features for the app. In view of this, we offer design recommendations for researchers and designers of sociocultural food recognition systems.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| 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) | 0.002 | 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 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".