Research
Explore my scientific research and publications in astronomy, data science, and machine learning. From stellar classification to interstellar pattern recognition, discover the methodologies and findings of ongoing investigations.
Explore my scientific research and publications in astronomy, data science, and machine learning. From stellar classification to interstellar pattern recognition, discover the methodologies and findings of ongoing investigations.
Selected academic papers and publications across various scientific disciplines. These works have been peer-reviewed and published in respected journals and conference proceedings.
Journal of Astronomical Computing, 45(3), 211-228
This paper presents novel machine learning algorithms for classifying stellar objects based on their spectral characteristics. We demonstrate significant improvements in accuracy over traditional methods, particularly for M-type and variable stars.
Proceedings of the International Conference on Data Science in Astronomy, 156-172
In this paper, we introduce Project AXIOM, a collaborative framework combining citizen science with advanced AI tools to accelerate the discovery and classification of celestial objects. Initial results show a 40% increase in discovery rates.
Astrophysical Journal, 892(1), 45-62
We apply convolutional neural networks to identify subtle patterns in Hertzsprung-Russell diagrams that correlate with stellar evolution stages. Our approach identifies previously undetected relationships between spectral characteristics and age.
Journal of Visual Communication and Image Representation, 82, 103417
This paper proposes a new algorithm for converting images to ASCII art that optimizes for perceptual similarity rather than pixel-level matching. The resulting images show better preservation of visual features and details.
Active research initiatives currently in progress. These projects represent ongoing work that has not yet been formally published but shows promising preliminary results.
A comprehensive survey of deep sky objects using a combination of ground-based telescopes and space telescope data. This project aims to catalog and classify previously unidentified objects and create a public database.
Development of a predictive model that can forecast stellar evolution based on current spectroscopic data. The model uses time-series analysis and deep learning to predict changes over billions of years.
A multidisciplinary study analyzing the astronomical accuracy in famous artworks throughout history, with a special focus on Van Gogh's Starry Night and other celestial-inspired pieces.
Active partnerships with academic institutions, research organizations, and industry partners working together on scientific exploration and discovery.
With: International Astronomical Union, NASA Ames Research Center
A global partnership to standardize and democratize access to astronomical data from observatories worldwide. Project AXIOM contributes machine learning models for data processing and classification.
With: Tech University AI Lab, Quantum Computing Research Group
Developing specialized AI algorithms optimized for scientific data analysis, with a focus on pattern recognition in complex, high-dimensional datasets common in astronomy and physics.
With: National Gallery of Modern Art, Institute for Creative Studies
A unique collaboration exploring the intersection of scientific visualization and artistic expression, with a special focus on astronomical imagery and its interpretation through various artistic media.
Tools, datasets, and educational materials available for research purposes. These resources are designed to support both independent researchers and collaborative projects.
A curated dataset containing spectral data and classifications for over 100,000 stars, including metadata and confidence scores.
Access DatasetA specialized dataset for machine learning applications, featuring HR diagrams with labeled stellar classifications and evolutionary stages.
Access DatasetA Python library for spectral analysis and stellar classification using machine learning models trained on the AXIOM dataset.
GitHub RepositoryA tool for converting images to ASCII art with customizable parameters and optimization for perceptual similarity.
Online Tool Python PackageA comprehensive course covering the fundamentals of data science applied to astronomical problems, including code examples and exercises.
Access CourseA detailed guide to the research methodologies developed for Project AXIOM, including experimental design, data collection protocols, and analysis techniques.
Download PDF Interactive Version