Comparison of decision tree methods for finding active objects
Advances in Space ResearchPublished 27 July 2007Open access
Yongheng Zhao, Yanxia Zhang
Citations340
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Several kinds of decision trees for finding active objects by multi-wavelength data, such as REPTree, Random Tree, Decision Stump, Random Forest, J48, NBTree, AdTree are described and experimental results show that ADTree is the best only in terms of accuracy.
Abstract
The automated classification of objects from large catalogues or survey\nprojects is an important task in many astronomical surveys. Faced with various\nclassification algorithms, astronomers should select the method according to\ntheir requirements. Here we describe several kinds of decision trees for\nfinding active objects by multi-wavelength data, such as REPTree, Random Tree,\nDecision Stump, Random Forest, J48, NBTree, AdTree. All decision tree\napproaches investigated are in the WEKA package. The classification performance\nof the methods is presented. In the process of classification by decision tree\nmethods, the classification rules are easily obtained, moreover these rules are\nclear and easy to understand for astronomers. As a result, astronomers are\ninclined to prefer and apply them, thus know which attributes are important to\ndiscriminate celestial objects. The experimental results show that when various\ndecision trees are applied in discriminating active objects (quasars, BL Lac\nobjects and active galaxies) from non-active objects (stars and galaxies),\nADTree is the best only in terms of accuracy, Decision Stump is the best only\nconsidering speed, J48 is the optimal choice considering both accuracy and\nspeed.\n
Keywords
EngineeringPhysics and Astronomy
Machine LearningRandom Forests
126,192 Citations2001Leo Breiman
Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the forest, and are also applicable to regression.
Elsevier eBooksData Mining: Practical Machine Learning Tools and Techniques
25,718 Citations2011Ian H. Witten, Eibe Frank
Machine LearningInduction of Decision Trees
14,815 Citations1986J. R. Quinlan
This paper summarizes an approach to synthesizing decision trees that has been used in a variety of systems, and it describes one such system, ID3, in detail, which is described in detail.
Machine LearningInduction of decision trees
12,383 Citations1986J. R. Quinlan
This paper summarizes an approach to synthesizing decision trees that has been used in a variety of systems, and it describes one such system, ID3, in detail, which is described in detail.
Scaling up the accuracy of Naive-Bayes classifiers: a decision-tree hybrid
1,363 Citations1996Ron Kohavi
A new algorithm, NBTree, is proposed, which induces a hybrid of decision-tree classifiers and Naive-Bayes classifiers: the decision-Tree nodes contain univariate splits as regular decision-trees, but the leaves contain Naïve-Bayesian classifiers.
The Alternating Decision Tree Learning Algorithm
715 Citations1999Yoav Freund, Llew Mason
A new type of classi cation rule, the alternating decision tree, which is a generalization of decision trees, voted decision trees and voted decision stumps and generates rules that are usually smaller in size and thus easier to interpret.
Monthly Notices of the Royal Astronomical SocietyMorphological Classification of galaxies by Artificial Neural Networks
147 Citations1992Michael C. Storrie‐Lombardi, O. Lahav +2 more
It is shown that the neural network behaves in the problem as a Bayesian classifier, i.e. it assigns the a posteriori probability for each of the five classes considered in the catalogue, and the network highest probability choice agrees with the catalogue classification.
Publications of the Astronomical Society of the PacificEstimating Photometric Redshifts Using Support Vector Machines
124 Citations2005Yogesh Wadadekar
This work presents a new approach to obtaining photometric redshifts using a kernel learning technique called “support vector machines,” which is likely to produce results that are comparable to the best results obtained using template fitting and artificial neural networks.
The Astronomical JournalIdentifying Red Variables in the Northern Sky Variability Survey
105 Citations2004Szymon Woźniak, W. T. Vestrand +1 more
The Astrophysical JournalRobust Machine Learning Applied to Astronomical Data Sets. I. Star‐Galaxy Classification of the Sloan Digital Sky Survey DR3 Using Decision Trees
100 Citations2006Nicholas M. Ball, Robert J. Brunner +2 more
This work presents the first public release of objects classified in this way for an entire SDSS data release, and details selection criteria within the probability space defined by the three classes to extract samples of stars and galaxies to a given completeness and efficiency.
Monthly Notices of the Royal Astronomical Society LettersFast cosmological parameter estimation using neural networks
81 Citations2007T. Auld, M. Bridges +2 more
It is found that the average error in the power spectra is typically 2 − 3% of cosmic variance, and that COSMONET is ∼ 7 × 10 4 faster than CAMB (for flat models) and ∼ 6 × 10 6 times faster than the official WMAP3 likelihood code.
Solar PhysicsAutomatic Solar Flare Detection Using MLP, RBF, and SVM
78 Citations2003Ming Qu, Frank Y. Shih +2 more
Experimental results show that by using SVM the authors can obtain the best classification rate of the solar flares, and believe this work will lead to real-time solar-flare detection using advanced pattern recognition techniques.
Monthly Notices of the Royal Astronomical SocietyPhysical parametrization of stellar spectra: the neural network approach
74 Citations1997C. A. L. Bailer-Jones, M. Irwin +2 more
It is demonstrated through the metallicity dependence of the derived temperature calibration that the neural networks are sensitive to the meetallicity signature in the real spectra.
Elsevier eBooksSKICAT: A Machine Learning System for Automated Cataloging of Large Scale Sky Surveys
53 Citations1993Usama M. Fayyad, N. Weir +1 more
The application of decision tree based classification techniques to the development of an automated tool for the reduction of a large scientific data set is described and SKICAT represents a system in which machine learning played a powerful and enabling role, and solved a difficult, scientifically significant problem.
Astronomy and AstrophysicsAutomated clustering algorithms for classification of astronomical objects
48 Citations2004Y. Zhang, Y. Zhao
This study concludes that in the situation of fewer features, LVQ and SLP show better performance, in contrast to SVM, which shows better performance when considering more features.
Springer Link (Chiba Institute of Technology)Automated clustering algorithms for classification of astronomical objects
42 Citations2004Yong Zhang, Yongheng Zhao
Publications of the Astronomical Society of the PacificClassification in Multidimensional Parameter Space: Methods and Examples
40 Citations2003Yanxia Zhang, Yongheng Zhao
Vistas in AstronomySpectral classification with principal component analysis and artificial neural networks
33 Citations1994Michael C. Storrie‐Lombardi, M. Irwin +2 more
This work has explored the use of principal component analysis for data pre-processing prior to classification of stellar spectra with a non-linear neural network to significantly enhances classification replicability, network stability, and convergence.
The Astrophysical JournalAutomated Classification of<i>ROSAT</i>Sources Using Heterogeneous Multiwavelength Source Catalogs
31 Citations2004T. A. McGlynn, A. A. Suchkov +9 more
It is concluded that the ClassX classifiers yield reasonable and reliable classifications for ROSAT sources and have the potential to broaden class representation significantly for rare object types.
Vistas in AstronomyHubble classification of galaxies using neural networks
19 Citations1994Anthony I. Adams, Anthony Woolley
The results show that correlations exist between the morphological type of a galaxy and measurable galaxy features, indicating that it is possible to define the Hubble sequence in terms of specific galaxy parameters.
Chinese Journal of Astronomy and AstrophysicsA Comparison of BBN, ADTree and MLP in separating Quasars from Large Survey Catalogues
18 Citations2007Yanxia Zhang, Yongheng Zhao
These automated methods are robust and effective for classifying point sources and may all be applied to large survey projects and for other astronomical issues, such as the parameter measurement of stars and the redshift estimation of galaxies and quasars.
Vistas in AstronomyStar-galaxy separation using neural networks
16 Citations1994S. C. Odewahn, M. Nielsen
A two color survey of 9 fields of the first epoch Palomar Sky Survey, centered on the North Galactic Pole, is discussed, which has been performed with the Minnesota Automated Plate Scanner (APS).
Vistas in AstronomyUltraviolet stellar spectral classification using a multilevel tree neural network
12 Citations1994R. K. Gulati, Ranjan Gupta +2 more
A pattern classification technique based on an Artificial Neural Network in a multi-level tree configuration to classify ultraviolet stellar spectra from the IUE Low-Dispersion Spectra Reference Atlas and preliminary results show that 94% of the spectra have been classified correctly.
Automatic Solar Flare Detection Using MLP, RBF and SVM
4 Citations2003Mingyue Qu, Frank Y. Shih +2 more