Advances in Hydrographic Data Processing: Time for a Paradigm Shift
Bibliographic record
Abstract
With currently available software, hydrographic data processing can be challenging for even experienced users. Though software generally keeps up with advances in hardware and processing methods, many of the frustrations are due to the fact that the human operator must connect all the pieces together to come up with the final processing solution. The problem in many cases is the human operator. Humans make typing errors when transcribing vessel configurations from one software application to another or from one coordinate frame convention to another. Humans make errors when importing ancillary data and then failing to associate it with the correct data files. Humans make errors when changing processing configurations and not then triggering the appropriate reprocessing or perhaps not triggering it for the correct files. Any human error along the way leads to poor results in the final product and many wasted hours, days or even weeks troubleshooting the source of the error. A paradigm shift is a fundamental change in approach or underlying assumptions. With the release of QPS Qimera, we are striving for a paradigm shift in that we are automating the mundane and error prone tasks for which computers are well suited but humans are not. Examples of where computers beat human operators include data transcription, unit conversion, coordinate frame transposition, processing state management and job scheduling. Qimera isolates the stages for which a human brings value to the process. Examples where humans win out (for the moment) include data integration troubleshooting, processing configuration management and data validation. With these concepts in mind, Qimera was built from the ground up with a Guided Workflow design philosophy that lets non-expert users arrive at typical bathymetric deliverables with little training or expert knowledge. Qimera also provides a Dynamic Workflow in that it is easy to make processing configuration adjustments or to perform data validation and to immediately assess the impacts of changes. The shortened feedback cycle between cause and effect promotes causal reasoning, a key ingredient for natural human learning processes. In effect, it allows users to train themselves. In this talk, we examine common sources of human error and demonstrate potential improvements to workflows with working examples of Qimera in action. Presenter Bio Jonathan Beaudoin has a Ph.D (2010) in Geodesy and Geomatics Engineering from the University of New Brunswick and Bachelor's degrees in Geodesy and Geomatics Engineering (2002) and Computer Science (2002), also from UNB. After finishing his Ph.D, he came to CCOM and did research in the field of echosounding uncertainty associated with oceanographic variability, seabed backscatter processing and improving best practices in multibeam echosounder fleet management as the Principal Investigator of the NSF-funded Multibeam Advisory Committee. After nearly four years at CCOM, Jonathan returned to Fredericton, Canada in 2013 to work for QPS where he is Chief Scientist and Product Manager for FMGT, FM Midwater and Qimera.
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 | 0.071 | 0.056 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.005 | 0.016 |
| Scholarly communication | 0.016 | 0.037 |
| Open science | 0.005 | 0.013 |
| Research integrity | 0.009 | 0.027 |
| Insufficient payload (model declined to judge) | 0.016 | 0.011 |
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".