A comparison of top-down and bottom-up approaches to recognizing component assemblies in image mining electronic circuits
Bibliographic record
Abstract
Historians of electronics (and subjects that depend on electronics like communications, instrumentation, and computation) have access to a vast digitized archive of primary sources. The majority of these sources are freely available. Turkel and various collaborators have used web crawlers to collect millions of pages of these documents to subject to automated analysis with text and image mining. Electronic schematics and circuit diagrams, which are ubiquitous in these sources, can be approached as a special case of line drawing. They provide interesting opportunities and challenges for the use of computer vision and image mining in historical research. In this paper, we present work-in-progress comparing two approaches to the problem of contextualization, the creation of functional tools built on image mining. An electronic schematic shows the interconnection of various kinds of components. Automating the recognition of these components is one step in a workflow for handling schematics. Historians of electronics are typically interested in meaningful assemblies that occur above the level of individual components, however. This is analogous to text mining, which becomes more useful at levels of analysis above the individual character. In the bottom-up approach, we start by recognizing the graphical symbols for individual components as assemblies of graphical primitives like lines and arcs, then build assemblies from those components. In the top-down approach, we start by analysing the whole schematic into components and connections, then abstract away from the components to extract the connections and the nodes where they meet.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".