Modeling Urban Land Use Changes Using Support Vector Machines

Support Vector Machines (SVM) is a machine learning (ML) algorithm commonly applied to the classification of remotely sensing data and more recently for modeling land use changes. However, in most geospatial applications the current literature does not elaborate on specifications of the SVM method with respect to data sampling, attribute selection and optimal parameters choices. Therefore the main objective of this study is to present and investigate the SVM technique for modeling urban land use change. The SVM model building procedure is presented together with the detailed evaluation of the output results with respect to the choice of datasets, attributes and the change of SVM parameters. Geospatial datasets containing nine land use classes and spatial attributes for the Municipality of Zemun, Republic of Serbia were used for years 2001, 2003, 2007 and 2011. The Correlation‐based Feature Subset method, kappa coefficient, Area Under Receiver Operating Characteristic Curve (AUC) and kappa simulation were used to perform the model evaluation and compare the model outputs with the real land use datasets. The obtained results indicate that the SVM‐based models perform better when implementing balanced data sampling, reduced data sets to informative subsets of attributes and properly identify the optimal learning parameters.     

Application of the photoacoustic method for characterization of natural galena (PbS)

Electron transport properties of single crystal and polycrystalline natural mineral galena (PbS) samples from the Trep?a mine, Yugoslavia, were determined using the photoacoustic frequency transmission technique. Their thermal diffusivity (D _T≈0.16 × 10^?5 m² s^?1), the coefficient of diffusion (D between 0.15×10^?2 0.16×10^?2 m² s^?1) and lifetime of the excess carrier (τ≈35 μs and the front and rear recombination velocity (s _g≈65.5 m s^?1 and s _b≈66.4 m s^?1, respectively), were calculated by comparing the experimental results and the theoretical photoacoustic amplitude and phase signals. The lattice parameter obtained by X-ray work was a?=5.936?Å. The free carrier concentration of these single-crystal samples was measured using the Hall method (N?=?3×10¹⁸ cm^?3). Measurements of the optical reflectivity of the same samples, as a function of wavelength, in the infrared and far infrared ranges, were performed. In the far infrared range a free electron plasma frequency was observed and numerically analyzed, using the least-squares fitting procedure. The values of optical parameters were calculated and the value of the free carrier concentration obtained by the Hall method was confirmed.     

Frequency modified fourier transform and its application to asteroids

Recently a method has been suggested to analyze the chaotic behaviour of a conservative dynamical system by numerical analysis of the fundamental frequencies. Frequencies and amplitudes are determined step by step. As the frequencies are not generally orthogonal, a Gramm-Schmidt orthogonalization is made and for each new frequency the old amplitudes of previously determined frequencies are corrected. For a chaotic trajectory variations of the frequencies and amplitudes determined over different time periods are expected. The change of frequencies in such a calculation is a measure of the chaoticity of the trajectory. While amplitudes are corrected, the frequencies (once determined) are constant. We suggest here simple linear corrections of frequencies for the effect of other close frequencies. The improvement of frequency determination is demonstrated on a model case. This method is applied to the first fifty numbered asteroids.     

Comparison of analytical and numerical calculations of the first class periodic orbits

The analytic construction of the direct periodic orbits in the circular 3-body problem is given in explicit form to the linear terms in . It is shown to be in good agreement with the numerically found orbits for large values of .     

A Method to Determine the Solar Synodic Rotation Rate and the Height of Tracers

The dependence of the measured apparent synodic solar rotation rate on the height of the chosen tracer is studied. A significant error occurs if the rotation rate is determined by tracing the apparent position of an object above the photospheric level projected on the solar disc. The centre-to-limb variation of this error can be used to determine simultaneously the height of the object and the true synodic rotation rate. The apparent (projected) heliographic coordinates are presented as a function of the height of the traced object and the coordinates of its footpoint. The relations obtained provide an explicit expression for the apparent rotation rate as a function of the observed heliographic coordinates of the tracer, enabling an analytic least-squares fit expression to determine simultaneously the real synodic rotation rate and the height of the tracer.     

A comparison of Hα and soft X-ray characteristics of spotless and spot group flares

A comparative analysis of spotless and spot group flares recorded at Hvar and Kanzelhöhe Observatories during the 21st cycle of solar activity is presented. The rate of occurrence of two-ribbon flares was found to be significantly higher for the spotless flares. In comparison with spot group flares of corresponding H importance, the soft X-ray peak values have been systematically lower for the spotless flares. The highest peak values and the energy released in soft X-rays was found for flares with a H ribbon protruding over a major spot umbra. It was found that the effective plasma temperatures in spotless flares have been considerably lower than the temperatures in spot group flares.     

The oscillating loop prominence of July 17, 1981

Oscillatory motions of a loop prominence observed on July 17, 1981 are analysed. The oscillations were mainly horizontal, with a period of 8 min. Restoring force was a result of magnetic tension, and assuming a simplified magnetic field configuration the expression for frequencies of oscillations is derived and compared with the observations. Taking the observed period, the strength of the magnetic field permeating the prominence is estimated as 45 G. Finally, the stability of the prominence is discussed.