Quasar candidate selection and redshift estimation using J-PLUS DR3 data
Bibliographic record
Abstract
ABSTRACT A comprehensive and high-purity quasar candidate catalogue with precise redshift measurements is crucial for advancing quasar research and cosmology. In the era of extensive sky surveys, the efficient identification of quasars from large-scale data sets has become a significant challenge in modern astronomy. By cross-matching the J-PLUS DR3 data set with unWISE and numerous spectroscopic data sets with accurate classifications, we compiled a known sample of 740 562 sources, including 338 456 stars, 320 606 galaxies, and 81 500 quasars. Subsequently, we developed several classification models employing XGBoost, CatBoost, and deep learning techniques. Through optimization of feature selection and hyperparameter tuning for each model, we derived an optimal classification model. This model achieved an accuracy of 99 per cent, with the Precision and Recall for quasar detection reaching 98.20 per cent and 99.39 per cent, respectively. In parallel, we utilized the known quasar sample to train an optimal model for redshift estimation, achieving a mean squared error of 0.139. Finally, combining the optimal classification and regression models, we designed an efficient workflow for quasar candidate selection and redshift estimation. This process resulted in the identification of over 3 million quasar candidates with photometric redshifts from the J-PLUS DR3 data set. These candidates provide an invaluable input catalogue for subsequent observations by large-scale spectroscopic surveys, such as LAMOST, SDSS, DESI, or other ongoing efforts in this field.
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 | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".