KMITL Expo 2026 LogoKMITL 66th Anniversary Logo

Investigation variable star classification through light curve analysis using machine learning approach

Abstract

With the development of space technology, wide-field sky surveys using telescopes have expanded the range of new data available for time-domain astronomical research. Traditional data analysis methods can no longer respond quickly and accurately enough to the growing volume of data. Thus, classifying time-series data, such as light curves, has become a significant challenge in the era of big data. In modern times, analyzing light curves has become essential for using machine learning techniques to handle and filter through massive amounts of data. Machine learning algorithms can be divided into two categories: shallow learning and deep learning. Numerous researchers have proposed and developed a variety of algorithms for light curve classification. In this study, we experimented with Support Vector Machine (SVM) and XGBoost, which are shallow machine learning algorithms, as well as 1D-CNN and Long Short-Term Memory (LSTM), which are deep learning algorithms, which are branches of deep machine learning, to classify variable stars. The training and testing data used in this study were from the Optical Gravitational Lensing Experiment-III (OGLE-III), consisting of variable star data from the Large Magellanic Cloud (LMC), categorized into five main classes: Classical Cepheids, δ Scutis, eclipsing binaries, RR Lyrae stars, and Long-period variables. The results demonstrate the performance analysis of each machine learning algorithm type applied to light curve data, while also highlighting the accuracy and statistical metrics of the algorithms used in the experiments.

Objective

ในงานนี้เราได้เสนอการใช้อัลกอริทึมการเรียนรู้ของเครื่องที่ทำการแบ่งอัลกอริทึมได้เป็น 2 ประเภท คือ แบบตื้นและแบบลึกมาทดสอบประสิทธิภาพโดยแบบตื้นมีมีอัลกอริทึม Support Vector Machine (SVM) และ XGBoost แบบลึกมีอัลกอริทึม 1D-CNN และ Long Short-Term Memory (LSTM) เราพิจารณาข้อมูลการสังเกตที่ได้จากฐานข้อมูล Optical Gravitational Lensing Experiment-III (OGLE-III) ที่เป็นดาวแปรแสงในพื้นที่ Large Magellanic Cloud (LMC) ด้วยกล้องโทรทรรศน์ขนาด 1.3-m Warsaw ที่ติดตั้งที่หอดูดาวลาสคัมปานัส ประเทศชิลี ข้อมูลนี้ประกอบด้วยการสังเกตดาวแปรแสงมากกว่าหนึ่งแสนครั้งโดยพิจารณาจากกราฟแสง และใช้ข้อมูลสถิติต่างๆ เช่น Accuracy, Precision, Recall, F1-score, AUG, mPa, mcc และ kappa ซึ่งงานวิจัยนี้มีจุดมุ่งหมายเพื่อที่จะทดสอบประสิทธิภาพในการจำแนกประเภทของดาวแปรแสงโดยใช้ข้อมูลการวิเคราะห์ light curve ด้วยเทคนิคการเรียนรู้ของเครื่องทั้งสองประเภท เพื่อให้เห็นถึงความเข้าใจในลักษณะและพฤติกรรมของดาวแปรแสง ซึ่งใช้ในประโยชน์ต่างๆ เช่น ความรู้ในด้านดาราศาสตร์ฟิสิกส์หรือการค้นพบดาวเคราะห์ดวงใหม่ๆ และการป้องกันภัยจากดาวแปรแสงมีอาจจะมีผลกระทบต่อโลก อีกทั้งในเรื่องการประหยัดเวลาและทรัพยากรในการที่จะจำแนกประเภทดาวแปรแสงอย่างมีระบบและมีประสิทธิภาพ

Other Innovations

Smart Agricultural Rail Robotics System

วิทยาลัยนวัตกรรมการผลิตขั้นสูง

Smart Agricultural Rail Robotics System

Smart Agriculture has rapidly developed in recent years, particularly with the integration of robotics and automation technologies to improve production efficiency and reduce costs, thereby enhancing the quality of current agricultural practices. A key innovation in this area is the rail-based robotic arm, designed to enhance work efficiency using a rail system with high precision and effectiveness. The application of this robotic arm covers various processes, such as planting, sorting, maintenance, harvesting, and resource management, allowing continuous operation and reducing human labor in repetitive and high-risk tasks. Studies have shown that the use of rail-based robotic arms in agriculture can significantly improve work efficiency, reduce production costs, and effectively mitigate environmental impact. By using robots in agricultural processes, it is possible to reduce contamination, lower the risk of crop damage, and make agriculture more sustainable. Additionally, it can increase accuracy in operations on limited spaces or farms with diverse crops. From these findings, it can be concluded that adopting rail-based robotic arm technology in agriculture not only enhances long-term production efficiency but also promotes sustainable agriculture and maximizes resource use, meeting future agricultural demands

Read more
Selection of landrace rice varieties resistant to saline soil

คณะเทคโนโลยีการเกษตร

Selection of landrace rice varieties resistant to saline soil

Rice is a salt-sensitive crop. The objective of this study was to evaluate the effect of salinity at flowering stage on physiological traits and yield of landrace rice. The experiment design was 4*10 Factorial in RCBD with 4 replications. Factor A was four salinity levels: control, 6, 12 and 16 dS/m; Factor B was 10 rice varieties. Data were collected on physiological traits and grain yield. The results showed that increasing salinity level decreased rice yield. The highest yield reduction was found when the rice received salt stress at 16 dS/m. In addition, rice varieties showed different yield performance when exposed to salt stress. In this found that Hom Yai variety had the lowest yield reduction when grown at 16 dS/m salinity level and did not differ from salt tolerant check variety.

Read more
Mistake in the First Teleportation Experimental

คณะสถาปัตยกรรม ศิลปะและการออกแบบ

Mistake in the First Teleportation Experimental

The concept for this work came from my curiosity about what would happen if, during interdimensional travel in space, a teleportation system were used. This system involves removing matter from one point and transferring it to another while maintaining its original state. If an error occurs and the matter is recreated or fused together, it could result in an experimental creature merging with the spacecraft. I choose the tardigrade as the first experimental subject for teleportation because the water bear has already been sent into space and survived. Therefore, I thought that if we were to actually test this teleportation system, the tardigrade would likely be one of the creatures chosen for experimentation.

Read more