The problem of nonperforming loans is one of the biggest problems in the banking sector. In order to mitigate this problem, it is necessary to improve the methods of credit risk assessment. One way to minimize credit risk is to improve the assessment of the creditworthiness of the applicant. In order to make a more accurate assessment, many models have been developed using classification techniques. This paper demonstrates the use of classification techniques in the form of a single classifier or in a classifier ensemble setting. We proposed bagging as a model ensemble using artificial neural networks. In the experiment conducted with the Bosnian commercial banks dataset, the proposed model showed promising results according to evaluation criteria, especially after the process of feature selection. Both individual and wrapper feature selection methods were used. Bagging with neural network (NNBag) outperforms commonly used techniques with accuracy improvement from 1% to 5%. The superiority of the proposed model (NNBag) is confirmed on two widely available datasets for assessing creditworthiness. Based on experimental results on three datasets, it is proven that NNBag is suitable for use in the assessment of the creditworthiness of applicants.

Gender equality together with the empowerment of women and girls through Information and Communication Technologies (ICT) is one of the critical enablers of sustainable development. This article aims to give a comprehensive overview of the main international and EU (European Union) strategic documents focused on bridging the digital gender gap. Overview of activities toward digital gender equality in Bosnia and Herzegovina has been presented too.

Introduction: Schizophrenia(SCZ) and Bipolar disorder (BD) are frequently occurring and impairing disorders that affect around 1% of the population. Important endophenotypes in the genetic research of SCZ and BD are cognitive functions. Core symptoms for SCZ and BD are impairments in working memory, declarative memory and attention, all of which fulfill the criteria for an endophenotype. The FK506 Binding Protein 5 (FKBP5) gene codes for a co-chaperone of the glucocorticoid receptor and has been reported to be associated with cognition. Aim: The aims of our research were to determine the degree of cognitive impairment in patients suffering from SCZ and BD and to explore the association of the FKBP5 variant rs3800373 genotype with the cognitive endophenotypes. Material and Methods: Patients and healthy controls were recruited over a period of two years from the Psychiatric Clinic, Clinical Center University of Sarajevo. Genotyping and neuropsychological assessments were performed for 263 subjects (129 SCZ, 53 BD, and 81 healthy controls [HC]). Neuropsychological assessments were performed for all patients with the Trail Making Test-A&B (TMT-A&B) and Digit-span forward&backwards tasks. The single nucleotide polymorphism (SNP) rs3800373 in the FKBP5 gene was genotyped using Infinium PsychArray Bead Chips. Results and Conclusion: SCZ and BD patients performed lower than HC in the TMT-A&B and in the Digit-span backwards task, while no differences were observed between SCZ and BD patients. While SCZ patients performed lower than HC in the Digit-span forwards task, there were no differences between BD and HC or between BD and SCZ. Rs 3800373 was not associated with performance in the TMT-A&B or Digit-span forwards&backwards tasks. SCZ and BD share largely overlapping neurocognitive characteristics. Rs3800373 was not associated with performance in the neuropsychological tests. However, given the limited sample size, the results do not exclude an association with the rs3800373 variant in a larger sample. Furthermore, as the analysis was limited to one SNP, the results cannot be generalized to other genetic variants in FKBP5.

Abstract This article presents a new approach to detect coiled coil and leucine zipper (L-Zip) motifs in protein sequences. The approach is based on protein scale calculation and sequence analysis. For this purpose, the wavelet-based local extrema extraction is employed, and window-based variations of local extrema afterward. This, in turn, provided a way to distinguish coiled coil subsequences and potential L-Zip motifs. The approach is validated on carefully chosen protein sequences that return inconclusive results within known frameworks for L-Zip detection, for example, 2ZIP. The results show that this new approach represents an improvement over previously presented approaches.

Microcapsules are used in various fields of application, such as in pharmacy, medicine, agriculture, chemical industry, construction industry, food industry, biotechnology, electronics. Fragranced and PCM (phase change materials) microcapsules also found their use in the printing and the textile industries, where they are applied in the combination with the appropriate ink or varnish on the desired substrate material. Microcapsules are applied either by coating or by different printing techniques, which main advantage is the ability to transfer the microcapsules onto desired areas of the substrate material without or with as little damage as possible, thus allowing the deposited microcapsules to fulfil their basic functionality. The aim of this research was to investigate the morphologic characteristics of the fragranced microcapsules and the coated prints using selected varnish and different fragranced microcapsules concentrations, as well as to determine how variable concentrations of the applied microcapsules in the varnish affect the optical characteristics of the coated prints. Performed SEM (scanning electron microscopy) and spectrodensitometric analyses of the coated prints revealed that both the coating process without microcapsules, as well as the different fragranced microcapsules’ concentration in the coated water-based varnish, significantly affected both the morphologic and the optical characteristics of the coated prints.

The human body transforms the chemical energy of the food into the work and the heat through the process of metabolism. The produced heat through the skin is transferred to the environment. In this case, in the state of thermal equilibrium, the amount of heat produced is equal to the amount of heat lost by conduction, convection, radiation, evaporation and breathing. The process of conduction of heat is transferred from the body to the environment, through layers of clothing and air, with the person standing still. Conductivity of heat in clothes depends on the thermal conductivity of the fibers from which the clothes are made, the conductivity of the air trapped in the pores of the clothes and the air on the surface of the clothing, the surface of the clothing layer through which the heat and the thickness of the clothes pass. The amount of heat transferred by conduction is usually negligible because the clothing, by its characteristics, slows down heat transfer in this way. Additionally, ink layers made in printing process significantly affect the properties of textile materials and clothes made from these materials. And today textile materials are increasingly being subjected to the process of printing due to aesthetic requirements of the people. This paper investigates the influence of digital printing parameters on the thermo-physiological features of textile materials. The essential print parameter was a different number of passes. In this research were used textile fabric materials of 100% cotton fibers. With printing process parameters, such as number of passes in the print, it is possible to influence the amount of ink that is applied on and in printed material, and thus the achievement of desired values of thermal parameters of printed materials. The influence of print parameters to thermo-physiological properties of the material is evaluated through a thermal conductivity and heat retention ability. Results of the research demonstrated that, in addition to material composition, the printing process with its parameters have a significant influence on the thermo-physiological characteristics of textile materials. The values of the thermal conductivity of the printed samples show that the increase in the number of application of ink in the printing results in a rise in the value of thermal conductivity coefficient, and decrease in heat retention ability value.

Phthalate esters are suspected endocrine disruptors that are found in a wide range of applications. The aim of this study was to determine the excretion of urinary metabolites in 16 individuals after inhalation and/or dermal exposure to 100–300 µg/m3 of deuterium-labelled diethyl phthalate (D4-DEP) and bis(2-ethylhexyl) phthalate (D4-DEHP). Dermal exposure in this study represents a case with clean clothing acting as a barrier. After inhalation, D4-DEP and D4-DEHP metabolites were excreted rapidly, though inter-individual variation was high. D4-DEP excretion peaked 3.3 h (T½ of 2.1 h) after combined inhalation and dermal exposure, with total excreted metabolite levels ranging from 0.055 to 2.351 nmol/nmol/m3 (nmol of urinary metabolites per phthalates air concentration in (nmol/m3)). After dermal exposure to D4-DEP, metabolite excretion peaked 4.6 h (T½ of 2.7 h) after exposure, with excreted metabolite levels in between 0.017 and 0.223 nmol/nmol/m3. After combined inhalation and dermal exposure to D4-DEHP, the excretion of all five analysed metabolites peaked after 4.7 h on average (T½ of 4.8 h), and metabolite levels ranged from 0.072 to 1.105 nmol/nmol/m3 between participants. No dermal uptake of particle phase D4-DEHP was observed. In conclusion, the average excreted levels of metabolites after combined inhalation and dermal exposure to D4-DEP was three times higher than after combined exposure to D4-DEHP; and nine times higher than after dermal exposure of D4-DEP. This study was made possible due to the use of novel approaches, i.e., the use of labelled phthalate esters to avoid the background concentration, and innovative technique of phthalate generation, both in the particle and the gas phase.

Studying the behavior of social insect using computer vision algorithms is an interesting topic for both biological and signal processing communities. One of the most interesting aspects in the field is tracking of honeybees. Regarding computer vision method, honeybees’ behavior has been mostly monitored inside and at the entrance of the hive. In this research we are proposing the method for automatic monitoring of honeybees’ activity outside of the hive. Experiments showed that the activity of honeybees outside the hive can estimated using an ultra-high definition video captured with UAV from distance of 10 meters. Specific spots where honeybees are gathered can be detected using heat maps which represent the density of their occurrence in the observed time interval.

This paper investigates the importance of different parameters of split-brain autoencoder to performance of learned image representations for remote sensing scene classification. We investigate the usage of LAB color space as well as color space created using PCA applied to RGB pixel values. We show that these two spaces give almost equal results, with slight favor towards the LAB color space. We also investigate choices of different quantization methods of color targets and number of quantization bins. We have found that using k-means clustering for quantization works slightly better than using uniform quantization. We also show that even when using really small number of bins it is possible to get only slightly worse results.

This paper describes the preparation and application of “home-made” iodide ISE enriched with ZnO nanoparticles. Iodide ISE membrane was made of AgI:Ag 2 S:PTFE = 1:1:2 and enriched with ZnO nanoparticles in ratio 1-5.0 wt.%. Prepared membranes were used for penicillamine (Pen) determination in acetic buffer (pH = 4-4.75) and in perchloric acid (pH = 1-2). Due to the low price method, simplicity and relative speed, a possibility for Pen determination was found. Pen determination showed better results in an acetic buffer, pH = 4 than in perchloric acid. All experiments for Pen measurements were done without pretreatment of pharmaceuticals. The determination is based on the reaction between Pen with both Ag + and Zn 2+ from the electrode membrane. A newly described method has linear response range for Pen 2.45×10 −6 -1×10 −2 mol L −1 and a detection limit of 2.24×10 −6 mol L −1 . The found concentrations of Pen are in a very good agreement with the declared ones with the standard deviation of 2.70 %.

Rheumatoid arthritis (RA) is an inflammatory joint disease that eventually leads to permanent bone and cartilage destruction. Fas has already been established as the regulator of inflammation in RA, but its role in bone formation under arthritic conditions is not completely defined. The aim of this study was to assess the effect of Fas inactivation on the bone damage during murine antigen‐induced arthritis. Subchondral bone of wild‐type (WT) and Fas‐knockout (Fas−/−) mice was evaluated by histomorphometry and microcomputerized tomography. Proportions of synovial bone and cartilage progenitors were assessed by flow cytometry. Synovial bone and cartilage progenitors were purified by fluorescence‐activated cell sorting and expression of Fas and Fas‐induced apoptosis were analyzed in vitro. Results showed that Fas−/− mice developed attenuated arthritis characterized by preserved epiphyseal bone and cartilage. A proportion of the earliest CD200+ bone and cartilage progenitors was reduced in WT mice with arthritis and was unaltered in Fas−/− mice. During osteoblastic differentiation in vitro, CD200+ cells express the highest levels of Fas and are removed by Fas ligation. These results suggest that Fas‐induced apoptosis of early CD200+ osteoprogenitor population represents potential mechanism underlying the impaired bone formation in arthritis, so their preservation may represent the bone‐protective mechanism during arthritis.—Lazić Mosler, E., Lukač, N., Flegar, D., Fadljević, M., Radanović, I., Cvija, H., Kelava, T., Ivčević, S., Šućur, A., Markotić, A., Katavić, V., Marušić, A., Grčević, D., Kovačić, N. Fas receptor induces apoptosis of synovial bone and cartilage progenitor populations and promotes bone loss in antigen‐induced arthritis. FASEB J. 33, 3330–3342 (2019). www.fasebj.org