Data Augmentation and Class Imbalance Compensation Using CTGAN to Improve Gas Detection Systems
Bibliographic record
Abstract
The use of sensors in gas detection systems for environmental monitoring is largely affected by sensor drift over time which reduces accurate classification. This drift can be minimized by using machine learning models trained on sensor data. Here, two different machine learning models are trained on the Gas Sensor Array Drift Dataset. However, this dataset, which has been collected over three years, suffers not only from drift but also from class imbalance. As a result, machine learning models cannot perform properly on this dataset. To address these problems, this paper introduces an innovative methodology for data compensation and augmentation using Conditional Tabular Generative Adversarial Networks (CTGAN). By employing this methodology, we can counteract the class imbalance and limit drift by bringing diversity to the dataset, which in turn improves the accuracy of machine learning models for gas detection systems. With class imbalance compensation, Multi-Layer Perceptron (MLP) and Support Vector Machines (SVM) achieved an improvement in classification accuracy in five batches, up to 20% for certain batches. Through data augmentation, they reached higher accuracy across six batches, with certain batches exceeding a 10% improvement. These achievements highlight the effectiveness and reliability of the use of synthetic data generation in tabular data for sensors.
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 | 0.003 |
| 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 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| 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".