Image search at work? An AI pipeline to classify and geo-locate figures from corporate subsurface documents.
Bibliographic record
Abstract
Summary The presentation will focus on the application of world class artificial intelligence technologies to reveal and extract knowledge from ConocoPhillips datasets. For this project, conducted by Kadme and ConocoPhillips, on top of the Lumin platform, Kadme has entered a collaboration with Inmeta, a consulting company part of the Crayon group. One of the objectives of the project was to gain a better understanding of images contained within the unstructured data realm and being able to utilise the enriched data for a quicker understanding of their context. In fact, not all knowledge can be located using search terms and metadata. A domain expert can learn a lot about a document just by looking at the figures contained within it. The solution must be capable of scaling to 10s of millions of figures, from dozens of different sources. The model must also remain relevant based on feedback from the end users, so retraining workflows are required. Further, the solution and the pipeline must be repeatable in many different environments. These subsequent deployments may wish to use different figure classes, so a workflow for automated bootstrapping of a new model also must be created.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.018 | 0.017 |
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".