Two comprehensive lists of articles on the Self-Organizing Map (SOM) have been published earlier in the Neural Computing Surveys. They contain references to scientific papers that deal with analyses and applications of the SOM, or have essentially benefited from the SOM. The previous lists together contained 5384 papers from the years 1981-2001. The present addendum contains 2333 new articles on the SOM from the years 20022005. We have also provided a keyword index to help finding the articles of interest.

/pdf/bibliography-of-self-organizing-map-som-papers-2002-2005-45m90im19s.pdf

Bibliography of self-organizing map (som) papers: 2002-2005 addendum

We introduced a decision tree method called Random Forests for multi-wavelength data classification. The data were adopted from different databases, including the Sloan Digital Sky Survey (SDSS) Data Release five, USNO, FIRST and ROSAT. We then studied the discrimination of quasars from stars and the classification of quasars, stars and galaxies with the sample from optical and radio bands and with that from optical and X-ray bands. Moreover, feature selection and feature weighting based on Random Forests were investigated. The performances based on different input patterns were compared. The experimental results show that the random forest method is an effective method for astronomical object classification and can be applied to other classification problems faced in astronomy. In addition, Random Forests will show its superiorities due to its own merits, e.g. classification, feature selection, feature weighting as well as outlier detection.

https://iopscience.iop.org/article/10.1088/1674-4527/9/2/011/pdf

Random forest algorithm for classification of multiwavelength data

We calculate photometric redshifts from the Sloan Digital Sky Survey Data Release 2 (SDSS DR2) Galaxy Sample using artificial neural networks (ANNs). Different input sets based on various parameters (e.g. magnitude, color index, flux information) are explored. Mainly, parameters from broadband photometry are utilized and their performances in redshift prediction are compared. While any parameter may be easily incorporated in the input, our results indicate that using the dereddened magnitudes often produces more accurate photometric redshifts than using the Petrosian magnitudes or model magnitudes as input, but the model magnitudes are superior to the Petrosian magnitudes. Also, better performance results when more effective parameters are used in the training set. The method is tested on a sample of 79 346 galaxies from the SDSS DR2. When using 19 parameters based on the dereddened magnitudes, the rms error in redshift estimation is σz = 0.020184. The ANN is highly competitive tool compared to the traditional template-fitting methods when a large and representative training set is available.

Estimating Photometric Redshifts with Artificial Neural Networks and Multi-Parameters

The important step of data preprocessing of data mining is feature
selection. Feature selection is used to improve the performance of
data mining algorithms by removing the irrelevant and redundant
features. By positional cross-identification, the multi-wavelength
data of 1656 active galactic nuclei (AGNs), 3718 stars, and 173
galaxies are obtained from optical (USNO-A2.0), X-ray (ROSAT), and
infrared (Two Micron All- Sky Survey) bands. In this paper we
applied a kind of filter approach named ReliefF to select features
from the multi-wavelength data. Then we put forward the naive
Bayes classifier to classify the objects with the feature subsets
and compare the results with and without feature selection, and
those with and without adding weights to features. The result
shows that the naive Bayes classifier based on ReliefF algorithms
is robust and efficient to preselect AGN candidates.

An automated classification algorithm for multiwavelength data

Descending stairs and floors classification as control reference in autonomous smart wheelchair

The sizes of astronomical surveys in different wavebands are increasing rapidly. Therefore, automatic classification of objects is becoming ever more important. We explore the performance of learning vector quantization (LVQ) in classifying multi-wavelength data. Our analysis concentrates on separating active sources from non-active ones. Different classes of X-ray emitters populate distinct regions of a multidimensional parameter space. In order to explore the distribution of various objects in a multidimensional parameter space, we positionally cross-correlate the data of quasars, BL Lacs, active galaxies, stars and normal galaxies in the optical, X-ray and infrared bands. We then apply LVQ to classify them with the obtained data. Our results show that LVQ is an effective method for separating AGNs from stars and normal galaxies with multi-wavelength data.

Yan-Xia Zhang

Papers

Learning Vector Quantization for Classifying Astronomical Objects