Quadrature-Inspired Generalized Choquet Integral in an Application to Classification Problems
Bibliographic record
Abstract
Correct classification remains a challenge for researchers and practitioners developing algorithms. Even a minor enhancement in classification quality, for instance, can significantly boost the effectiveness of detecting conditions or anomalies in safety data. One solution to this challenge involves aggregating classification results. This process can be executed effectively as long as the aggregation function is appropriately chosen. One of the most efficient aggregation operators is the Choquet integral. Furthermore, there exist numerous generalizations and extensions of the Choquet integral in the existing literature. In this study, we conduct a comprehensive analysis and evaluation of a novel approach for deriving an aggregate classification. The aggregation process applied to various classifiers is based on enhancements to the Choquet integral. These novel expressions draw inspiration from Newton-Cotes quadratures and other well-known formulae from numerical analysis. In contrast to previous approaches that exploit the generalization of the Choquet integral, our approach requires the utilization of two or three adjacent values associated with the membership of a specific element in different classes. This enables the use of more efficient enhancements in terms of accuracy measurement. Specifically, the t-norm following the integral symbol can be effectively replaced by mathematical expressions used in executing numerical integration formulae. This leads to more precise results and aligns with the concept of numerical integration. Furthermore, in a series of experiments, we thoroughly assess the performance of the proposed approach in terms of classification accuracy. We analyze the strengths and weaknesses of the new approach and establish the experimental settings that can be applied to similar tasks. In the series of experiments, we have demonstrated that the proposed Quadrature-Inspired Generalized Choquet Integral (QIGCI) can either outperform previous enhancements of the Choquet integral or at least achieve a similar level of accuracy measurement. However, we also highlight scenarios where previous approaches can still be a suitable choice. The number of QIGCI-based aggregation models that outperform others is convincing, indicating that this approach is worthy of consideration.
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 | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".