Validation of scientific topic models using graph analysis and corpus metadata

Probabilistic topic modeling algorithms like Latent Dirichlet Allocation (LDA) have become powerful tools for the analysis of large collections of documents (such as papers, projects, or funding applications) in science, technology an innovation (STI) policy design and monitoring. However, selecting an appropriate and stable topic model for a specific application (by adjusting the hyperparameters of the algorithm) is not a trivial problem. Common validation metrics like coherence or perplexity, which are focused on the quality of topics, are not a good fit in applications where the quality of the document similarity relations inferred from the topic model is especially relevant. Relying on graph analysis techniques, the aim of our work is to state a new methodology for the selection of hyperparameters which is specifically oriented to optimize the similarity metrics emanating from the topic model. In order to do this, we propose two graph metrics: the first measures the variability of the similarity graphs that result from different runs of the algorithm for a fixed value of the hyperparameters, while the second metric measures the alignment between the graph derived from the LDA model and another obtained using metadata available for the corresponding corpus. Through experiments on various corpora related to STI, it is shown that the proposed metrics provide relevant indicators to select the number of topics and build persistent topic models that are consistent with the metadata. Their use, which can be extended to other topic models beyond LDA, could facilitate the systematic adoption of this kind of techniques in STI policy analysis and design.

     

Magnetically Modified Waste Bark of Aegle marmelos Tree as a Promising Biosorbent for Uranium(VI) Sorption

Radiochemistry - Magnetically modified waste bark of the Aegle marmelos tree was prepared by using green synthesis method and was used in a batch system for U(VI) removal from aqueous solution. The...     

Justification for the Use of Cohen’s Kappa Statistic in Experimental Studies of NLP and Text Mining

Modern metrics for evaluating agreement coefficients between the experimental results and expert opinion are compared, and the possibility of using these metrics in experimental research in automatic text processing by machine learning methods is assessed. The choice of Cohen’s kappa coefficient as a measure of expert opinion agreement in the NLP and Text Mining problems is justified. An example of using Cohen’s kappa coefficient for evaluating the level of agreement between the opinion of an expert and the results of ML classification and the measure of agreement of expert opinions in the alignment of sentences of the Kazakh-Russian parallel corpus is given. Based on this analysis, it is proved that Cohen’s kappa coefficient is one of the best statistical methods for determining the level of agreement in experimental studies due to its ease of use, computing simplicity, and high accuracy of the results.

     

Motion of a Robot Mower with Directional Control

Russian Engineering Research - The stability of robot-mower motion in a specific direction is considered. The direction is regulated by means of an angular sensor and a programmable controller...     

Quantitative Analysis of Valley–Orbit Coupling in Germanium Doped with Group-V Donors

Semiconductors - The wave functions of electrons localized at P, As, and Sb shallow donors in Ge are calculated in the envelope function approximation taking into account valley–orbit...     

Influence of pH on the DSSC Performance of Template Assisted SnO2 Nanostructures

Journal of Inorganic and Organometallic Polymers and Materials - This paper reports the performance of alizarin based SnO2 dye-sensitized solar cells. A novel template assisted method has been...     

Phase formation,structure and properties of light-weight aluminosilicate proppants based on clay-diabase and clay-granite binary mixes

One of the drawbacks of fusible clays is the narrow sintering interval due to a sharp increase in the amount of iron-silicate melt at a temperature of 1000–1100 °C, which hardens in the form of a glass phase upon cooling. This leads to a relatively low mechanical strength of the calcined samples and causes the danger of melting the granular material surface from such clays during the firing process. To increase the strength of samples of fusible clays, the influence of diabase and granitoid rocks was considered. It was found that the strengthening effect of diabase and granitoid rock additives in an amount of 20–50% in a mixture with fusible clay is due to an increase of total content of the crystalline phase (mullite, cristobalite and residual quartz) from 18–20% in clays without additives to 22–28 % - in mixtures with diabase and to 28–34% - with granitoid additives) at a temperature of 1050–1100 °C. This increase is due to the activation of synthesis processes of secondary mullite and crystallization from alkali-rich feldspar melt of amorphous silica, released from the structure of clay minerals. The established influence of the igneous rocks used made it possible to develop compositions and propose process flow sheet for producing aluminosilicate proppants based on fusible clays. The use of granitoid and diabase rocks in an amount of 20–70% with fusible clays produces lightweight aluminosilicate proppants with bulk density of 1.40–1.46 g/cm³ at temperature range of 1050–1100 °C, which can endure destructive pressures up to 34.5–52 MPa.     

Some Molecular and Cellular Stress Mechanisms Associated with Neurodegenerative Diseases and Atherosclerosis

Chronic stress is a combination of nonspecific adaptive reactions of the body to the influence of various adverse stress factors which disrupt its homeostasis, and it is also a corresponding state of the organism’s nervous system (or the body in general). We hypothesized that chronic stress may be one of the causes occurence of several molecular and cellular types of stress. We analyzed literary sources and considered most of these types of stress in our review article. We examined genes and mutations of nuclear and mitochondrial genomes and also molecular variants which lead to various types of stress. The end result of chronic stress can be metabolic disturbance in humans and animals, leading to accumulation of reactive oxygen species (ROS), oxidative stress, energy deficiency in cells (due to a decrease in ATP synthesis) and mitochondrial dysfunction. These changes can last for the lifetime and lead to severe pathologies, including neurodegenerative diseases and atherosclerosis. The analysis of literature allowed us to conclude that under the influence of chronic stress, metabolism in the human body can be disrupted, mutations of the mitochondrial and nuclear genome and dysfunction of cells and their compartments can occur. As a result of these processes, oxidative, genotoxic, and cellular stress can occur. Therefore, chronic stress can be one of the causes forthe occurrence and development of neurodegenerative diseases and atherosclerosis. In particular, chronic stress can play a large role in the occurrence and development of oxidative, genotoxic, and cellular types of stress.     

Multicomponent Diamond-Like Semiconductors Based on the InBV–CdS System: Bulk and Surface Properties

Semiconductors - According to developed methods, in the fields of the mutual solubility of initial binary compounds (InP, InSb, and CdS), solid solutions of the InP–CdS and InSb–CdS...     

Impact behavior of aminosilane functionalized nanosilica based shear thickening fluid impregnated Kevlar fabrics

In the present work, two types of shear thickening fluids have been synthesized by using neat and aminosilane functionalized silica nanoparticles and their viscosity curves have been obtained by the rheometer. Based on the values of peak viscosity of synthesized shear thickening fluids, the surface functionalized nanosilica based shear thickening fluid has been chosen as a best candidate due to the high viscosity for impregnation into the neat Kevlar of different layers viz. four (04) and eight (08) layers for velocity impact study. The experimental investigations reveal high energy absorption of shear thickening fluid impregnated Kevlar as compared to the neat Kevlar. The maximum energy absorption 62 J is achieved corresponding to the initial velocity 154 m∙s⁻¹ for 08 layers shear thickening fluid impregnated Kevlar specimen. The data have also been analytically determined and validated with the experimental data. The experimental data have good agreement with the analytical data within the accuracy of around 15 to 20%. The present findings can have significant inferences towards the fabrication of shear thickening fluids using nanomaterials for numerous applications such as soft armors, dampers, nanofinishing and so forth.