• Animal Bioscience
• /
• v.34 no.4
• /
• pp.770-782
• /
• 2021

Objective: The aim of the present study was to compare the effectiveness of three approaches (the seasonal auto-regressive integrated moving average [SARIMA] model, the nonlinear autoregressive exogenous [NARX] artificial neural networks and Wood's model) to the prediction of milk yield during lactation. Methods: The dataset comprised monthly test-day records from 965 Polish Holstein-Friesian Black-and-White primiparous cows. The milk yields from cows in their first lactation (from 5 to 305 days in milk) were used. Each lactation was divided into ten lactation stages of approximately 30 days. Two age groups and four calving seasons were distinguished. The records collected between 2009 and 2015 were used for model fitting and those from 2016 for the verification of predictive performance. Results: No significant differences between the predicted and the real values were found. The predictions generated by SARIMA were slightly more accurate, although they did not differ significantly from those produced by the NARX and Wood's models. SARIMA had a slightly better performance, especially in the initial periods, whereas the NARX and Wood's models in the later ones. Conclusion: The use of SARIMA was more time-consuming than that of NARX and Wood's model. The application of the SARIMA, NARX and Wood's models (after their implementation in a user-friendly software) may allow farmers to estimate milk yield of cows that begin production for the first time.

• Nutrition Research and Practice
• /
• v.15 no.2
• /
• pp.160-172
• /
• 2021

BACKGROUND/OBJECTIVES: Nutritional status and food intake during pregnancy and lactation can affect fetal programming. In the current metabolic syndrome epidemic, high-fructose diets have been strongly implicated. This study investigated the effect of maternal high-fructose intake during pregnancy and lactation on the development of metabolic syndrome in adult offspring. SUBJECTS/METHODS: Drinking water with or without 20% fructose was administered to female C57BL/6J mice over the course of their pregnancy and lactation periods. After weaning, pups ate regular chow. Accu-Chek Performa was used to measure glucose levels, and a tail-cuff method was used to examine systolic blood pressure. Animals were sacrificed at 7 months, their livers were excised, and sections were stained with Oil Red O and hematoxylin and eosin (H&E) staining. Kidneys were collected for gene expression analysis using quantitative real-time Polymerase chain reaction. RESULTS: Adult offspring exposed to maternal high-fructose intake during pregnancy and lactation presented with heavier body weights, fattier livers, and broader areas under the curve in glucose tolerance test values than control offspring. Serum levels of alanine aminotransferase, aspartate aminotransferase, glucose, triglycerides, and total cholesterol and systolic blood pressure in the maternal high-fructose group were higher than that in controls. However, there were no significant differences in mRNA expressions of renin-angiotensin-aldosterone system genes and sodium transporter genes. CONCLUSIONS: These results suggest that maternal high-fructose intake during pregnancy and lactation induces metabolic syndrome with hyperglycemia, hypertension, and dyslipidemia in adult offspring.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.673-686
• /
• 2022

Deep learning algorithms for Structural Health Monitoring (SHM) have been extracting the interest of researchers and engineers. These algorithms commonly used loss functions and evaluation indices like the mean square error (MSE) which were not originally designed for SHM problems. An updated loss function which was specifically constructed for deep-learning-based structural damage detection problems has been proposed in this study. By tuning the coefficients of the loss function, the weights for damage localization and quantification can be adapted to the real situation and the deep learning network can avoid unnecessary iterations on damage localization and focus on the damage severity identification. To prove efficiency of the proposed method, structural damage detection using convolutional neural networks (CNNs) was conducted on a truss bridge model. Results showed that the validation curve with the updated loss function converged faster than the traditional MSE. Data augmentation was conducted to improve the anti-noise ability of the proposed method. For reducing the training time, the normalized modal strain energy change (NMSEC) was extracted, and the principal component analysis (PCA) was adopted for dimension reduction. The results showed that the training time was reduced by 90% and the damage identification accuracy could also have a slight increase. Furthermore, the effect of different modes and elements on the training dataset was also analyzed. The proposed method could greatly improve the performance for structural damage detection on both the training time and detection accuracy.

• Communications for Statistical Applications and Methods
• /
• v.29 no.5
• /
• pp.499-512
• /
• 2022

Intervention measures have been implemented worldwide to reduce the spread of the COVID-19 outbreak. The COVID-19 outbreak has occured in several waves of infection, so this paper is divided into three groups, namely those countries who have passed the pandemic period, those countries who are still experiencing a single-wave pandemic, and those countries who are experiencing a multi-wave pandemic. The purpose of this study is to develop a multi-wave Richards model with several changepoint detection methods so as to obtain more accurate prediction results, especially for the multi-wave case. We investigated epidemiological trends in different countries from January 2020 to October 2021 to determine the temporal changes during the epidemic with respect to the intervention strategy used. In this article, we adjust the daily cumulative epidemiological data for COVID-19 using the logistic growth model and the multi-wave Richards curve development model. The changepoint detection methods used include the interpolation method, the Pruned Exact Linear Time (PELT) method, and the Binary Segmentation (BS) method. The results of the analysis using 9 countries show that the Richards model development can be used to analyze multi-wave data using changepoint detection so that the initial data used for prediction on the last wave can be determined precisely. The changepoint used is the coincident changepoint generated by the PELT and BS methods. The interpolation method is only used to find out how many pandemic waves have occurred in given a country. Several waves have been identified and can better describe the data. Our results can find the peak of the pandemic and when it will end in each country, both for a single-wave pandemic and a multi-wave pandemic.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.177-189
• /
• 2023

Malware detection is an increasingly important operational focus in cyber security, particularly given the fast pace of such threats (e.g., new malware variants introduced every day). There has been great interest in exploring the use of machine learning techniques in automating and enhancing the effectiveness of malware detection and analysis. In this paper, we present a deep recurrent neural network solution as a stacked Long Short-Term Memory (LSTM) with a pre-training as a regularization method to avoid random network initialization. In our proposal, we use global and short dependencies of the inputs. With pre-training, we avoid random initialization and are able to improve the accuracy and robustness of malware threat hunting. The proposed method speeds up the convergence (in comparison to stacked LSTM) by reducing the length of malware OpCode or bytecode sequences. Hence, the complexity of our final method is reduced. This leads to better accuracy, higher Mattews Correlation Coefficients (MCC), and Area Under the Curve (AUC) in comparison to a standard LSTM with similar detection time. Our proposed method can be applied in real-time malware threat hunting, particularly for safety critical systems such as eHealth or Internet of Military of Things where poor convergence of the model could lead to catastrophic consequences. We evaluate the effectiveness of our proposed method on Windows, Ransomware, Internet of Things (IoT), and Android malware datasets using both static and dynamic analysis. For the IoT malware detection, we also present a comparative summary of the performance on an IoT-specific dataset of our proposed method and the standard stacked LSTM method. More specifically, of our proposed method achieves an accuracy of 99.1% in detecting IoT malware samples, with AUC of 0.985, and MCC of 0.95; thus, outperforming standard LSTM based methods in these key metrics.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.2030-2052
• /
• 2023

Analysis of a bipartite (two-mode) network is a significant research area to understand the formation of social communities, economic systems, drug side effect topology, etc. in complex information systems. Most of the previous works talk about a projection-based model or latent feature model, which predicts the link based on singular similarity. The projection-based models suffer from the loss of structural information in the projected network and the latent feature is hardly present. This work proposes a novel method for link prediction in the bipartite network based on an ensemble of composite similarities, overcoming the issues of model-based and latent feature models. The proposed method analyzes the structure, neighborhood nodes as well as latent attributes between the nodes to predict the link in the network. To illustrate the proposed method, experiments are performed with five real-world data sets and compared with various state-of-art link prediction methods and it is inferred that this method outperforms with ~3% to ~9% higher using area under the precision-recall curve (AUC-PR) measure. This work holds great significance in the study of biological networks, e-commerce networks, complex web-based systems, networks of drug binding, enzyme protein, and other related networks in understanding the formation of such complex networks. Further, this study helps in link prediction and its usability for different purposes ranging from building intelligent systems to providing services in big data and web-based systems.

• Earthquakes and Structures
• /
• v.25 no.3
• /
• pp.199-208
• /
• 2023

Seismic performance analysis is one of the fundamental steps in the design of new or retrofitting buildings. In the seismic performance analysis, the adapted spectral acceleration curve for a given site mainly governs the seismic behavior of buildings. Since every soil site (class) has a different impact on the spectral accelerations of input motions, different spectral acceleration curves have to be involved for every soil class that the building is located on top of. Modern seismic design codes (e.g., Eurocode 8, EC8, or Turkish Building Earthquake Code, TBEC) provide design response spectra for all the soil classes to be used in the building design or retrofitting. This research aims to evaluate the EC8 and TBEC based design response spectra using the spectra of real earthquake input motions that occurred (and were recorded at only soil classes A, B and C, no recording is available at soil class D) in a specific area in Turkey. It also conducts response spectrum analyses of 5, 10 and 13 floor reinforced concrete building models under EC8, TBEC and actual spectral response curves. The results indicate that the EC8 and especially TBEC given design response spectra cannot be able to represent the mean actual spectral acceleration curves at soil classes A, B and C. This is particularly observed at periods higher than 0.3 s, 0.42 s and 0.55 s for the TBEC design response spectra, 0.54 s, 0.65 s and 0.84 s for the EC8 design response spectra at soil classes A, B and C, respectively. This is also reflected to the shear forces of three building models, as actual spectral acceleration curves lead to the highest shear forces, followed by the shear forces obtained from EC8 and, then, the TBEC design response spectra.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.687-697
• /
• 2023

Botnets, whose attack patterns are becoming more sophisticated and diverse, are recognized as one of the most serious cybersecurity threats today. This paper revisits the experimental results of botnet detection using autoencoder, a semi-supervised deep learning model, for UGR and CTU-13 data sets. To prepare the input vectors of autoencoder, we create data points by grouping the NetFlow records into sliding windows based on source IP address and aggregating them to form features. In particular, we discover a simple power-law; that is the number of data points that have some flow-degree is proportional to the number of NetFlow records aggregated in them. Moreover, we show that our power-law fits the real data very well resulting in correlation coefficients of 97% or higher. We also show that this power-law has an impact on the learning of autoencoder and, as a result, influences the performance of botnet detection. Furthermore, we evaluate the performance of autoencoder using the area under the Receiver Operating Characteristic (ROC) curve.

• Analytical Science and Technology
• /
• v.36 no.5
• /
• pp.216-223
• /
• 2023

Aluminium (Al) is one of the major elements in rocks and its concentration can be varied, depending on different rock types as well as sources. The present study aimed to propose an analytical method based on the UV-Vis as a cheap, simple, and common instrument equipped in most laboratories for Al quantification in rocks after the microwave assisted acid digestion. The aluminone and 8-hydroxyquinoline were investigated for the colorimetric assay. The results show that the 8-hydroxyquinoline reagent was more favorable in terms of the minimized affects of the potential interferences present in the digested solutions, i.e., Fe³⁺, Si⁴⁺ and F^-. The calibration curve was constructed from 0.10 mg/L to 3.00 mg/L with the goodness of linearity (R² = 0.9996). The limits of detection and quantification (LOD and LOQ) were estimated, i.e., 0.029 mg/L and 0.087 mg/L, respectively. The 8-hydroxyquinoline was applied to real rock samples, demonstrating favorable precision (RSD = 0.34 %-1.8 %) and no remarkable differences were found compared to the inductively coupled plasma-mass spectrometry (ICP-MS) as a reference measurement approach.

• Journal of the Korean Society of Radiology
• /
• v.17 no.6
• /
• pp.897-902
• /
• 2023

The purpose of this paper is to analyze the characteristics of Silicon Photomultiplier (SiPM) for the realization of high-sensitivity radiation detection in portable detectors. Portable X-ray detectors offer the advantage of quickly accessing the patient's location and obtaining real-time images, allowing physicians to perform rapid diagnoses. However, this mobility comes with challenges in achieving accurate radiation detection. In existing detectors, SiPM is used for a simple purpose of detecting X-ray triggers. To verify the feasibility of high-sensitivity X-ray detection through SiPM, seven types of SiPM sensors were compared and selected, and their characteristics were analyzed. The SiPM used in the final test demonstrated the ability to distinguish signals at the ultra-low radiation level of 10 nGy, and it was observed that the slope of the signal rise curve varies with the X-ray tube voltage. Utilizing the characteristics of SiPM, which exhibits changes in signal level and duration with X-ray dose, it appears possible to achieve high-sensitivity measurements for X-ray detection.