Stefano Cavuoti

The detection of globular clusters in galaxies as a data mining problem

ABSTRACT We present an application of self-adaptive supervised learning classifiers derived fromthe Machine Learning paradigm, to the identification of candidate Globular Clus-ters in deep, wide-field, single band HST images. Several methods provided by theDAME (Data Mining & Exploration) web application, were tested and compared onthe NGC1399 HST data described in Paolillo et al. (2011). The best results were ob-tained using a Multi LayerPerceptronwith Quasi Newton learningrule which achieveda classification accuracy of 98.3%, with a completeness of 97.8% and 1.6% contami-nation. An extensive set of experiments revealed that the use of accurate structuralparameters (effective radius, central surface brightness) does improve the final result,but only by ∼5%. It is also shown that the method is capable to retrieve also extremesources (for instance, very extended objects) which are missed by more traditionalapproaches.Key words: Globular clusters; elliptical galaxies; NGC1399; Machine Learning

Genetic Algorithm Modeling with GPU Parallel Computing Technology

We present a multi-purpose genetic algorithm, designed and implemented with GPGPU / CUDA parallel computing technology. The model was derived from a multi-core CPU serial implementation, named GAME, already scientifically successfully tested and validated on astrophysical massive data classification problems, through a web application resource (DAMEWARE), specialized in data mining based on Machine Learning paradigms. Since genetic algorithms are inherently parallel, the GPGPU computing paradigm has provided an exploit of the internal training features of the model, permitting a strong optimization in terms of processing performances and scalability.

Extracting Knowledge from Massive Astronomical Data Sets

The exponential growth of astronomical data collected by both ground-based and spaceborne instruments has fostered the growth of astroinformatics: a new discipline lying at the intersection between astronomy, applied computer science, and information and computation technologies. At the very heart of astroinformatics is a complex set of methodologies usually called data mining (DM) or knowledge discovery in databases (KDD). In the astronomical domain, DM/KDD are still in a very early usage stage, even though new methods and tools are being continuously deployed to cope with the massive data sets (MDSs) that can only grow in the future. In this paper, we briefly outline some general problems encountered when applying DM/KDD methods to astrophysical problems and describe the DAME (Data Mining and Exploration) Web application. While specifically tailored to work on MDSs, DAME can be effectively applied also to smaller data sets. As an illustration, we describe two applications of DAME to two different problems: the identification of candidate GCs in external galaxies and the classification of active Galactic nuclei (AGN). We believe that tools and services of this nature will become increasingly necessary for data-intensive astronomy (and indeed all sciences) in the twenty-first century.

DAME: A WEB ORIENTED INFRASTRUCTURE FOR SCIENTIFIC DATA MINING AND EXPLORATION

Nowadays, many scientific areas share the same need of being able to deal with massive and distributed datasets and to perform on them complex knowledge extraction tasks. This simple consideration is behind the international efforts to build virtual organizations such as, for instance, the Virtual Observatory (VObs). DAME (DAta Mining & Exploration) is an innovative, general purpose, Web-based, VObs compliant, distributed data mining infrastructure specialized in Massive Data Sets exploration with machine learning methods. Initially fine tuned to deal with astronomical data only, DAME has evolved in a general purpose platform which has found applications also in other domains of human endeavor. We present the products and a short outline of a science case, together with a detailed description of main features available in the beta release of the web application now released.

Data-rich astronomy: mining synoptic sky surveys

In the last decade a new generation of telescopes and sensors has allowed the production of a very large amount of data and astronomy has become, a data-rich science; this transition is often labeled as: data revolution and data tsunami. The first locution puts emphasis on the expectations of the astronomers while the second stresses, instead, the dramatic problem arising from this large amount of data: which is no longer computable with traditional approaches to data storage, data reduction and data analysis. In a new, age new instruments are necessary, as it happened in the Bronze age when mankind left the old instruments made out of stone to adopt the new, better ones made with bronze. Everything changed, even the social structure. In a similar way, this new age of Astronomy calls for a new generation of tools and, for a new methodological approach to many problems, and for the acquisition of new skills. The attems to find a solution to this problems falls under the umbrella of a new discipline which originated by the intersection of astronomy, statistics and computer science: Astroinformatics, (Borne, 2009; Djorgovski et al., 2006).

DAME: A Distributed Data Mining and Exploration Framework Within the Virtual Observatory

Nowadays, many scientific areas share the same broad requirements of being able to deal with massive and distributed datasets while, when possible, being integrated with services and applications. In order to solve the growing gap between the incremental generation of data and our understanding of it, it is required to know how to access, retrieve, analyze, mine and integrate data from disparate sources. One of the fundamental aspects of any new generation of data mining software tool or package which really wants to become a service for the community is the possibility to use it within complex workflows which each user can fine tune in order to match the specific demands of his scientific goal. These workflows need often to access different resources (data, providers, computing facilities and packages) and require a strict interoperability on (at least) the client side. The project DAME (data mining and exploration) arises from these requirements by providing a distributed Web-based data mining infrastructure specialized on massive data sets exploration with soft computing methods. Originally designed to deal with astrophysical use cases, where first scientific application examples have demonstrated its effectiveness, the DAME Suite results as a multi-disciplinary platform-independent tool perfectly compliant with modern KDD (knowledge discovery in databases) requirements and Information and Communication Technology trends.

Neural Gas Based Classification of Globular Clusters.

Within scientific and real life problems, classification is a typical case of extremely complex tasks in data-driven scenarios, especially if approached with traditional techniques. Machine Learning supervised and unsupervised paradigms, providing self-adaptive and semi-automatic methods, are able to navigate into large volumes of data characterized by a multi-dimensional parameter space, thus representing an ideal method to disentangle classes of objects in a reliable and efficient way. In Astrophysics, the identification of candidate Globular Clusters through deep, wide-field, single band images, is one of such cases where self-adaptive methods demonstrated a high performance and reliability. Here we experimented some variants of the known Neural Gas model, exploring both supervised and unsupervised paradigms of Machine Learning for the classification of Globular Clusters. Main scope of this work was to verify the possibility to improve the computational efficiency of the methods to solve complex data-driven problems, by exploiting the parallel programming with GPU framework. By using the astrophysical playground, the goal was to scientifically validate such kind of models for further applications extended to other contexts.

Data mining and knowledge discovery resources for astronomy in the web 2.0 age

The emerging field of AstroInformatics, while on the one hand appears crucial to face the technological challenges, on the other is opening new exciting perspectives for new astronomical discoveries through the implementation of advanced data mining procedures. The complexity of astronomical data and the variety of scientific problems, however, call for innovative algorithms and methods as well as for an extreme usage of ICT technologies. The DAME (DAta Mining and Exploration) Program exposes a series of web-based services to perform scientific investigation on astronomical massive data sets. The engineering design and requirements, driving its development since the beginning of the project, are projected towards a new paradigm of Web based resources, which reflect the final goal to become a prototype of an efficient data mining framework in the data-centric era.