Luminis Stellarum et Machina: Applications of Machine Learning in Light Curve Analysis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908539149090816 |
|---|---|
| author | Akhmetali, Almat Zhunuskanov, Alisher Sakan, Aknur Zaidyn, Marat Namazbayev, Timur Turlykozhayeva, Dana Ussipov, Nurzhan |
| author_facet | Akhmetali, Almat Zhunuskanov, Alisher Sakan, Aknur Zaidyn, Marat Namazbayev, Timur Turlykozhayeva, Dana Ussipov, Nurzhan |
| contents | The rapid advancement of observational capabilities in astronomy has led to an exponential growth in the volume of light curve (LC) data, creating both opportunities and challenges for time-domain astronomy. Traditional analytical methods often struggle to fully extract the scientific value of these large and complex datasets. Machine learning (ML) algorithms are increasingly used for LC analysis, enabling classification, prediction, pattern discovery and anomaly detection. However, research in this area remains fragmented, with no comprehensive synthesis of how ML methods address the specific challenges of LC data. Key difficulties include class imbalance, noisy or sparse measurements, effective feature extraction, and limited interpretability of models. This lack of a unified overview makes it difficult for researchers to identify suitable approaches or recognize unresolved problems that require methodological advances. To address this gap, this survey systematically reviews ML techniques applied to LC analysis, outlining their principles and applications in tasks such as exoplanet detection, variable star classification and supernova identification. By clarifying the current state of the field and highlighting open challenges, this work provides guidance for future research and supports the effective integration of ML into astronomical big data studies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_10038 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Luminis Stellarum et Machina: Applications of Machine Learning in Light Curve Analysis Akhmetali, Almat Zhunuskanov, Alisher Sakan, Aknur Zaidyn, Marat Namazbayev, Timur Turlykozhayeva, Dana Ussipov, Nurzhan Instrumentation and Methods for Astrophysics Solar and Stellar Astrophysics The rapid advancement of observational capabilities in astronomy has led to an exponential growth in the volume of light curve (LC) data, creating both opportunities and challenges for time-domain astronomy. Traditional analytical methods often struggle to fully extract the scientific value of these large and complex datasets. Machine learning (ML) algorithms are increasingly used for LC analysis, enabling classification, prediction, pattern discovery and anomaly detection. However, research in this area remains fragmented, with no comprehensive synthesis of how ML methods address the specific challenges of LC data. Key difficulties include class imbalance, noisy or sparse measurements, effective feature extraction, and limited interpretability of models. This lack of a unified overview makes it difficult for researchers to identify suitable approaches or recognize unresolved problems that require methodological advances. To address this gap, this survey systematically reviews ML techniques applied to LC analysis, outlining their principles and applications in tasks such as exoplanet detection, variable star classification and supernova identification. By clarifying the current state of the field and highlighting open challenges, this work provides guidance for future research and supports the effective integration of ML into astronomical big data studies. |
| title | Luminis Stellarum et Machina: Applications of Machine Learning in Light Curve Analysis |
| topic | Instrumentation and Methods for Astrophysics Solar and Stellar Astrophysics |
| url | https://arxiv.org/abs/2504.10038 |