SOPHIA: an interactive cluster-based retrieval system for the OHSUMED collection.

The ability to perform an exploratory search and retrieval of relevant documents from a large collection of domain-specific documents is an important requirement both in the field of medicine and other areas. In this paper, we present a unsupervised distributional clustering technique called SOPHIA. SOPHIA provides a semantically meaningful visual clustering of the document corpus in conjunction with an intuitive interactive search facility. We assess the effectiveness of SOPHIA's cluster-based information retrieval for the MEDLINE testset collection known as OHSUMED.     

Graphite Oxides Obtained from Porous Graphite: The Role of Surface Chemistry and Texture in Ammonia Retention at Ambient Conditions

Graphite oxides (GO) synthesized using Brodie and Hummers methods are tested for ammonia adsorption at ambient conditions with different contents of water in the system. Surface characterization before and after exposure to ammonia is performed using XRD, FTIR spectroscopy, potentiometric titration, thermal analysis, adsorption of nitrogen, and XPS. Oxidation of the same porous graphite using two methods results in materials with different textural and chemical features. On GO obtained using the Brodie method mainly epoxy and carboxylic groups are present whereas on the GO obtained using the Hummers method chemisorbed oxygen is also found. The contribution of the carboxylic groups in the latter material is greater. It also contains sulfur either in sulfones or as residual sulfuric acid. Ammonia is adsorbed either via reaction with surface groups or dissolution in water. The former is responsible for strong adsorption. The evidence of the catalytic effect of the carbon surface on activation of oxygen leading to surface oxidation is also observed.     

Mössbauer spectroscopic studies of the mineralogical changes in coal as a function of cleaning,pyrolysis, combustion and coalconversion processes

The mineralogical changes in a Perry County, Illinois coal from the Herrin (No. 6) Member due to cleaning, pyrolysis, combustion, and coal-conversion processes were studied. Mössbauer spectroscopy was used in tandem with X-ray diffraction to follow the changes in the forms of iron originally present in the coal resulting from processing. The chemistry of the pyrite conversion is less complex than expected. Iron does not become uniformly distributed in all possible minerals but tends to form simple products. Pyrrhotites along with spinel and hydrated ferrous sulphates are the primary mineral products found in coat liquefaction and pyrolysis process residues; while mullite, ferrous silicates and the iron oxides (hematite, geothite and magnetite) are the most abundant mineral products found in Lurgi gasification and power plant fly ashes. The detailed distribution of iron, however, is dependent upon conditions in the particular process equipment in which the coal is used and the conversion process in which it is used.     

Mössbauer studies of iron in Lurgi gasification ashes and power plant fly and bottom ash

Iron Mössbauer spectroscopy and X-ray diffraction methods were applied to the study of a selection of ashes produced in a Lurgi gasification plant and fly ash from a pulverized coal combustion. The ashes contained hematite, magnetite, and goethite. Sixty percent or more of the iron in these ashes was in the oxide form, with the remainder present in mullite and other silicate phases. Iron was divalent in the latter, and present as both Fe⁺² and Fe⁺³ in mullite. Ratios of Fe⁺² and Fe⁺³ varied from 0.3 to 0.7. By comparison, a water-quenched molten bottom ash was free of iron oxides and contained only amorphous silicate phases with virtually all iron in the divalent state.     

One-pot preparation of functionalized nanostructured carbons

Functionalized nanostructured carbons were obtained by template method using sucrose as carbon precursor and silica gel SG60 as structure directing agent. To introduce acid groups into carbon a different amount of phosphoric acid was added to sucrose (impregnation ratio 0–1) before filling the pores of silica gel. After carbonization at 800 °C silica template was removed by dissolution in HF. Carbons were characterized by SEM/EDX, nitrogen adsorption at 77 K and acid–base titration. Pore size distributions (PSD) indicated that carbons obtained at impregnation (IR) of 0–0.75 have pores with size less than 4 nm due to crack during carbonization and activating action of phosphoric acid. Templated pores with 5.1 nm are observed in all carbons. PSD significantly changed at IR = 1 showing additional large mesopores of 8.2 nm. Acid–base titration revealed very high concentration of acid surface groups (4.1–4.7 mmol/g), significant amount of which (46 ± 3%) belongs to phosphate groups (pK = 2.3 ± 0.1). Another surface groups are weak carboxylic (pK = 5.9 ± 0.3), lactone and/or enol (pK = 7.2 ± 0.2, pK = 8.8 ± 0.3) and phenol groups (pK = 10.5 ± 0.3). SEM showed smooth carbon surface up to impregnation ratio 0.5 and increasing roughness at higher amount of phosphoric acid.     

Adsorption of hydrogen sulfide on graphite derived materials modified by incorporation of nitrogen

Adsorbents for hydrogen sulfide were prepared by heat treatment at 950 °C of spent graphite oxide based materials previously used as ammonia removal media. The materials were characterized using adsorption of nitrogen, potentiometric titration, thermal analysis, FTIR and XRD. The dynamic adsorption experiments showed an enhanced adsorption capacity owing to the presence of basic nitrogen functionalities. Although on the surface of adsorbents hydrogen sulfide was mainly oxidized to elemental sulfur small amount of SO₂ was also formed. The removal capacity was found to depend on the dispersion of nitrogen containing species, which is governed by the porosity and structure of the adsorbents. The amount of nitrogen incorporated depends on the amount of ammonia on the surface and on the surface reactivity. The latter is determined by the amount of oxygen containing groups and the level of structural heterogeneity.     

Ultrashort pulse propagation in uniform and nonuniform waveguide long-period gratings

We conduct a detailed theoretical analysis of ultrashort pulse propagation through waveguide long-period grating (LPG) structures operating in the linear regime. We first consider the case of uniform LPGs, and we also investigate the effect of the typical grating nonuniformities, e.g., grating profile apodization, grating period chirping, and discrete phase shifts, on the spectral and temporal behavior of LPG structures. The two interacting modes are analyzed separately, and advanced representation tools, namely, space-wavelength and space-time diagrams (where space refers to the longitudinal grating dimension), are used to provide a deeper insight into the physics that determines the pulse evolution dynamics through the grating structures under analysis. In addition to its intrinsic physical interest, our study reveals the strong potential of LPG-based devices for optical pulse reshaping operations in the subpicosecond regime.     

Deep Semisupervised Learning-Based Network Anomaly Detection in Heterogeneous Information Systems

The extensive proliferation of modern information services and ubiquitous digitization of society have raised cybersecurity challenges to new levels. With the massive number of connected devices, opportunities for potential network attacks are nearly unlimited. An additional problem is that many low-cost devices are not equipped with effective security protection so that they are easily hacked and applied within a network of bots (botnet) to perform distributed denial of service (DDoS) attacks. In this paper, we propose a novel intrusion detection system (IDS) based on deep learning that aims to identify suspicious behavior in modern heterogeneous information systems. The proposed approach is based on a deep recurrent autoencoder that learns time series of normal network behavior and detects notable network anomalies. An additional feature of the proposed IDS is that it is trained with an optimized dataset, where the number of features is reduced by 94% without classification accuracy loss. Thus, the proposed IDS remains stable in response to slight system perturbations, which do not represent network anomalies. The proposed approach is evaluated under different simulation scenarios and provides a 99% detection accuracy over known datasets while reducing the training time by an order of magnitude.     

Effect of Al$$_$$O$$_3$$3 Nanoparticles Additives on the Density,Saturated Vapor Pressure,Surface Tension and Viscosity of Isopropyl Alcohol

This paper presents results of an experimental study of the density, saturated vapor pressure, surface tension and viscosity of Al\(_2\)O\(_3\) nanoparticle colloidal solutions in isopropyl alcohol. Studies of the thermophysical properties of nanofluids were performed at various temperatures and concentrations of Al\(_2\)O\(_3\) nanoparticles. The paper gives considerable attention to a turbidimetric analysis of the stability of nanofluid samples. Samples of nanofluids remained stable over the range of parameters of the experiments, ensuring the reliability of the thermophysical property data for the Al\(_2\)O\(_3\) nanoparticle colloidal solutions in isopropyl alcohol. The studies show that the addition of Al\(_2\)O\(_3\) nanoparticles leads to an increase of the density, saturated vapor pressure and viscosity, as well as a decrease for the surface tension of isopropyl alcohol. The information reported in this paper on the various thermophysical properties for the isopropyl alcohol/Al\(_2\)O\(_3\) nanoparticle model system is useful for the development of thermodynamically consistent models for predicting properties of nanofluids and correct modeling of the heat exchange processes.