Latest Dataset Research Articles

Significant of Blood Glucose Time-Series Forecasting with Temporal Fusion Transformer Model for Type 1 Diabetes

Blood glucose (BG) prediction has advanced to a state of the art thanks to deep learning models, which have been demonstrated to enhance type 1 diabetes (T1D) therapy. Nevertheless, most current models are limited to single-horizon prediction and have several practical drawbacks, including poor interpretability. In this study, we develop a novel approach to optimize the forecasting model in the training processes using the Optuna function, cross-validation, and the latest dataset. We offer a novel deep learning framework for multi-horizon BG prediction: the temporal fusion transformer (TFT). TFT employs a self-attention mechanism to extract long-term temporal dependencies and enables a model with an auto-tuning adjustment approach on hyperparameters and a cross-validation function on univariate and multivariate input models. On a clinical dataset with T1D subjects, namely ShanghaiT1DM (16 subjects), D1NAMO (9 subjects), and OhioT1DM (6 subjects) datasets, it achieved an average root mean square error (RMSE) of the three datasets used. The TFT model gave a lower error score than the baseline models of neural hierarchical interpolation for time series (N-HiTS)and long short-term memory (LSTM) with the values of 10.08+0.31 mg/dL and 12.34+0.62 mg/dL on the univariate input model and 9.18+1.21 mg/dL and 14.33+0.52 mg/dL on the multivariate input model for the prediction horizons of 30 and 60 minutes, respectively. This result explained that the TFT model was adequate for carrying out multi-horizon BG value-level forecasting and potentially deploying on-edge devices to improve clinical efforts to manage BG levels for T1D patients in real-time application.

Swin-chart: An efficient approach for chart classification

Charts are a visualization tool used in scientific documents to facilitate easy comprehension of complex relationships underlying data and experiments. Researchers use various chart types to convey scientific information, so the problem of data extraction and subsequent chart understanding becomes very challenging. Many studies have been taken up in the literature to address the problem of chart mining, whose motivation is to facilitate the editing of existing charts, carry out extrapolative studies, and provide a deeper understanding of the underlying data. The first step towards chart understanding is chart classification, for which traditional ML and CNN-based deep learning models have been used in the literature. In this paper, we propose Swin-Chart, a Swin transformer-based deep learning approach for chart classification, which generalizes well across multiple datasets with a wide range of chart categories. Swin-Chart comprises a pre-trained Swin Transformer, a finetuning component, and a weight averaging component. The proposed approach is tested on a five-chart image benchmark dataset. We observed that the Swin-Chart model outperformers existing state-of-the-art models on all the datasets. Furthermore, we also provide an ablation study of the Swin-Chart model with all five datasets to understand the importance of various sub-parts such as the back-bone Swin transformer model, the value of several best weights selected for the weight averaging component, and the presence of the weight averaging component itself.The Swin-Chart model also received first position in the chart classification task on the latest dataset in the CHART Infographics competition at ICDAR 2023 - chartinfo.github.io.

GaitGMT: Global feature mapping transformer for gait recognition

Gait recognition is an important biometric technology that allows for the remote collection of stakeholders’ characteristics, without requiring their explicit cooperation. It has gained considerable attention in the fields of criminal investigation and intelligent security. Previous studies have shown that local gait features can enhance gait recognition performance by improving robustness to disturbances. However, global gait features also play a crucial role in gait recognition. Many researchers have utilized convolutional operations to extract global features, but these operations tend to focus on features within the receptive field, neglecting those outside of it. Therefore, the potential of global gait features has not been fully explored. In this paper, we propose a gait recognition framework based on vision transformers, aiming to enhance the extraction of global gait features. We introduce an adaptive multi-frame global feature mapping (AMGM) method to address the challenge of inconsistent feature dimensions caused by variations in the number of gait frames when fusing global and local features. We evaluate our model on the latest datasets, and the experimental results demonstrate a significant breakthrough. Notably, our model achieves state-of-the-art recognition accuracy, particularly in scenarios where subjects are wearing coats. Additionally, our model achieves remarkable improvements in recognition accuracy through training with small sample sets.

Deep learning enabled intrusion detection system for Industrial IOT environment

The prevalence of security vulnerabilities in Internet of Things (IoT) applications poses a serious threat to enterprise systems, necessitating sophisticated and reliable defense solutions to counter emerging and evolving threats. For the Industrial Internet of Things (IIoT), stakeholders require trustworthy and sustainable systems that can prevent the loss of human life during critical operations. The impact of multi-variant persistent and sophisticated bot attacks on connected IIoTs is potentially catastrophic, and their detection presents a highly complex and critical challenge. Therefore, there is a pressing need for efficient and timely detection of IIoT botnet attacks. This research paper proposes a robust deep learning model named AttackNet for the detection and classification of different botnet attacks in IIoT based on adaptive based CNN-GRU model. The model is extensively evaluated using the latest dataset and standard performance evaluation metrics, demonstrating its capacity to protect IIoT networks against sophisticated cyber-attacks with a testing accuracy of 99.75%, a loss of 0.0063, precision and recall score of 99.75% and 99.74% respectively. Our proposed model demonstrates superior accuracy, particularly within the N_BaIoT dataset. It achieves an outstanding accuracy of 99.75% across ten classes, surpassing state-of-the-art techniques by a substantial margin ranging from 3.2% to 16.07%. Moreover, the proposed model outperforms state-of-the-art anomaly detection systems in IIoT based on a real-time IoT device dataset in terms of detecting and classifying botnet attacks accurately.

The role of the Foreign Direct Investment inflows on export in Azerbaijan: An ARDL approach

Purpose ― This study aims to investigate the impact of FDI inflows on Trade in Azerbaijan from 1993 to 2021. Method ― This study uses the datasets from the World Bank Database. It employs the Augmented Dickey and Fuller (ADF), Phillips and Perron (PP), Zivot and Andrews (ZA), ARDL bounds testing approach, and the Granger Causality tests for the empirical part of the study.Findings ― The bound test shows the presence of cointegration between FDI and Export. The estimated long-run equation suggests a positive and significant relationship, whereas the estimated short-run equation indicates a positive but insignificant relationship between FDI and export. Additionally, the results of Granger causality test show a unidirectional causality running from FDI to export. Implications ― Since the FDI inflows show a positive effect on the export of Azerbaijan, most foreign investments come into the oil and gas sector. Accordingly, oil and gas products and services account for a significant share of exports in Azerbaijan. Policymakers might need new regulations to attract more attention from foreign investors to non-oil sectors.Originality/value ― There were vast studies about FDI and trade relationships in different countries with different techniques. This study is unique because it employs a new methodology and the latest dataset in which Azerbaijan was a focused area for the first time.

Headache disorders and relevant sex and socioeconomic patterns in adolescents and young adults across 204 countries and territories: an updated global analysis.

Primary headache disorders are a group of highly prevalent and disabling neurological diseases that mainly consist of migraine and tension-type headache (TTH). A previous study showed that the burden of headaches peaked at a working age that ranged from 15 to 49, particularly among females, affecting their productivity and severely damaging their social interactions. The latest dataset was retrieved from the Global Burden of Disease (GBD) Study 2019. Three indicators, including prevalence, incidence, and years lived with disability (YLDs), were adopted for evaluation. The overall and specific headache burdens were fully compared and analysed at global, regional, and national levels. The ratio of female YLD rates to male YLD rates due to headaches was calculated to estimate the sex pattern. Finally, we utilized the two-tailed Spearman test to explore the potential association between socioeconomic background and headaches among young people. Globally, for overall headache disorders, a total of 2,049,979,883 prevalent cases (95% uncertainty interval (UI): 1,864,148,110 to 2,239,388,034), 601,229,802 incident cases (95% UI: 530,329,914 to 681,007,934), and 38,355,993 YLDs (95% UI: 7,259,286 to 83,634,503) were observed for those aged 10 to 54 in 2019. Sex differences were widely found for all headache types among adolescents and young adults, especially migraine. However, the most interesting finding was that the associations we tested between the socioeconomic environment and young headache patients were positive, regardless of region or specific country or territory. Overall, the global burden of headaches in adolescents and young adults largely increased from 1990 to 2019. Although slight declines were observed in sex differences, they remained significant and challenging. The positive correlations between headache and socioeconomic background among young people were relatively inconsistent with previous investigations, and several related hypotheses were proposed for explanation. Interdisciplinary actions involving education, policy- and law-making, and basic medical practice are desperately needed to further fight against the headache burden, promote gender equality in headache care, and eliminate the stigmatization of headache patients in student and working groups.

The Removal of Pollutants from Wastewater Using Magnetic Biochar: A Scientometric and Visualization Analysis.

In recent years, the use of magnetic biochar in wastewater treatment has shown significant effects and attracted scholars' attention. However, due to the relatively short research time and the lack of systematic summaries, it is difficult to provide a more in-depth analysis. This study utilizes RStudio and CiteSpace software to comprehensively analyze the research trends and progress of magnetic biochar in wastewater treatment. The analysis of bibliometrics is performed on 551 relevant papers retrieved from the Web of Science, spanning the period between 2011 and 2022. The most influential countries, institutions, journals, disciplinary distribution, and top 10 authors and papers in this field have been identified. The latest dataset has been used for keyword clustering and burst analysis. The results indicated that: (1) Bin Gao is the most influential author in this field, and high-level journals such as Bioresource Technology are more inclined to publish articles in the field of magnetic biochar. (2) Research in this field has predominantly focused on the removal of heavy metals and organic compounds. Keyword burst analysis shows a shift in research direction towards the removal of complex organic pollutants recently. (3) For the future development of magnetic biochar, an environment-friendly approach, economic viability, and joint technology are the directions that need more exploration. Finally, this paper provides a summary of the various adsorption mechanisms of magnetic biochar and several common modification methods, aiming to assist scholars in their research endeavors.

The Post-COVID-19 Economic Recovery, Government Performance and Lived Poverty Conditions in Kenya

The post-COVID-19 era has witnessed intensive engagements on latest datasets on COVID-19 experiences and the need for robust data systems to understand emerging policy pathways. There have also been extensive deliberations on how public organisations can deliver beyond immediate concerns with economic recovery and improved livelihoods. This paper draws insights from the Afrobarometer round 9 dataset on how different African governments' economies have been recovering from COVID-19 and how this has affected citizens' well-being in Kenya. The results show that despite the good recovery progress, the government has yet to stabilise high food prices, inflation, debts, job creation, etc. While national policy priorities may explain how this situation plays out, these challenges remain complex and linked to external factors that African governments may have little control over.

Inflationary interpretation of the stochastic gravitational wave background signal detected by pulsar timing array experiments

Various pulsar timing array (PTA) experiments (NANOGrav, EPTA, PPTA, CPTA, including data from InPTA) very recently reported evidence for excess red common-spectrum signals in their latest datasets, with inter-pulsar correlations following the Hellings-Downs pattern, pointing to a stochastic gravitational wave background (SGWB) origin. Focusing for concreteness on the NANOGrav signal (given that all signals are in good agreement between each other), I inspect whether it supports an inflationary SGWB explanation, finding that such an interpretation calls for an extremely blue tensor spectrum, with spectral index nT≃1.8±0.3, while Big Bang Nucleosynthesis limits require a very low reheating scale, Trh≲10GeV. While not impossible, an inflationary origin for the PTA signal is barely tenable: within well-motivated inflationary models it is hard to achieve such a blue tilt, whereas models who do tend to predict sizeable non-Gaussianities, excluded by observations. Intriguingly, ekpyrotic models naturally predict a SGWB with spectral index nT=2, although with an amplitude too suppressed to be able to explain the signal detected by PTA experiments. Finally, I provide explicit expressions for a bivariate Gaussian approximation to the joint posterior distribution for the intrinsic-noise amplitude and spectral index of the NANOGrav signal, which can facilitate extending similar analyses to different theoretical signals.

Spatial and Temporal Characteristics of Drought Events in Southwest China over the Past 120 Years

Global climate change, especially extreme drought events, presents a complicated challenge to humanity and Earth’s system in the 21st century. As an extremely important carbon sink region in China, Southwest China has encountered frequent drought disasters in recent decades. It is critical to explore the frequency, duration, severity, and other associated characteristics of drought events as well as their spatial and temporal patterns in the region from a long-term perspective. In this study, we used the latest dataset from the Spanish National Research Council (CSIC) between 1901 and 2018 to extract all drought events by calculating the standardized anomaly of the Standardized Precipitation Evapotranspiration Index (SPEI). Theil–Sen median trend analysis, the Mann–Kendall test, and the moving t-test were used to reveal the spatial trend and mutation point of drought severity. The results showed that (1) The standardized anomaly of the 3-month SPEI can accurately identify drought events in Southwest China. In total, 72 drought events occurred during this period, of which the consecutive drought in autumn, winter, and spring from 2009 to 2010 lasted the longest, having the most substantial severity and the most extensive damage range. (2) Drought events mainly started in spring and early summer and ended in autumn and winter. The distribution of drought was the most expansive and the drought severity was the most serious in September. (3) In terms of spatial pattern, Guangxi has the highest frequency of drought events, with some areas experiencing up to 100 events. The average duration of drought events ranged between 3.5 and 5.5 months, with most lasting for 4–5 months. The most severe drought areas are mainly concentrated in southern Sichuan and western Yunnan. Overall, the severity of drought events in the west were generally higher compared to that in the east. (4) Over the past 120 years, most of the region (82.46%) showed an increasing trend in drought severity, with a slope of up to −0.01. About 15.12% of the areas exhibited a significant drying trend (p < 0.05), particularly in southern Sichuan, eastern Guizhou, and northern and southern Yunnan. Such analyses can serve as a scientific foundation for developing drought prevention and mitigation measures as well as exploring how drought events affect the structure and function of terrestrial ecosystems in Southwest China.

Sleep Apnea and Substance Use Disorders Associated with Co-Occurrence of Anxiety Disorder and Depression among U.S. Adults: Findings from the NSDUH 2008–2014

Few studies have focused on sleep apnea and substance use disorders with co-occurrence of anxiety disorder and depression. This study included a total of 270,227 adults, 9268 with co-occurrence of anxiety disorder and depression in the past year, from the combined 2008-2014 National Survey on Drug Use and Health (NSDUH) data, which are the latest datasets with measures of anxiety disorder and sleep apnea. Weighted multinomial logistic regression analyses were used to estimate the associations between anxiety disorder and depression and their co-occurrence. Comorbidity was highly prevalent: 40.4% of those with depression also met the criteria for anxiety disorder, whereas 51.8% of those with anxiety disorder also met the criteria for depression. The prevalences of anxiety only and co-occurrence increased from 2008 to 2014. The prevalences of anxiety disorder only, depression only, and co-occurrence of anxiety disorder and depression in individuals with sleep apnea were 4.4%, 12.9%, and 12.2%, respectively, and the prevalences in substance use disorders were 6.4%, 9.4%, and 10.7%, respectively. The results showed that sleep apnea, substance use disorders, and nicotine dependence were significantly associated with increased odds of anxiety disorder, depression, and co-occurrence (all p values < 0.0001). Furthermore, several chronic diseases (asthma, bronchitis, hypertension, and heart disease) were associated with the co-occurrence of anxiety disorder and depression. These findings suggest clinicians and other healthcare providers consider screening for depression and anxiety with sleep apnea and substance use disorders for improved therapeutic outcomes.

Revisiting the environmental impact of renewable energy, non-renewable energy, remittances, and economic growth: CO2 emissions versus ecological footprint for top remittance-receiving countries.

The study examined the impact of renewable energy, non-renewable energy, remittances, and economic growth on environmental degradation from the perspective of carbon dioxide emissions (CO2) and ecological footprint for the top 50 remittance-receiving countries for 1991-2018. This study simulates the environmental future for attaining the targets of Sustainable Development Goal 7 (SDG7) using the latest datasets. This study is one of the few that empirically explores how various explanatory variables affect CO2 and ecological footprint. The study employed the pool mean group autoregressive distributive lag (PMG-ARDL), fully modified ordinary least squares (FMOLS), and dynamic ordinary least squares (DOLS) techniques for the analysis. In the long-run, non-renewable energy and economic growth have a positive impact, whereas renewable energy and remittances have a negative impact on CO2 and ecological footprint. The impact of non-renewable energy on CO2 and ecological footprint is more significant than renewable energy in both the short- and long-run. Most of the variables have a bidirectional causality among each other. This highlights the need for a paradigm shift towards renewable energy in the top recipients specifically for developing countries.

Validity of Environmental Kuznets Curve in the Malaysian Economy: A Fresh Evidence

It is acknowledged that environmental degradation, climate change and the atmosphere's temperature are increasing due to the emission of greenhouse gasses. Therefore, a speedy response is required to lessen the severe effects of greenhouse gases. The present study has analyzed the influence of electric consumption, fiscal development, globalization, and trade openness on carbon emission using the latest dataset and modern econometric techniques. To accomplish the said objectives, the study has employed Johansen cointegration, autoregressive distributed lag model (ARDK), and block exogeneity. The results specify that with the 1% increase in electric consumption, fiscal development, globalization and trade openness, the carbon dioxide emission increases by 0.01%, 0.04%, 0.68%, and 0.25%, respectively. Moreover, the results also indicated that the environmental Kuznets curve is prevailing in the Malaysian economy because an inverted U-shape connection occurs between carbon and GDP2. On the basis of data diagnostics, it is determined that the ARDL model is credible, stable, and reliable. Further, block exogeneity affirms the long-term affiliation among the factors and concludes that all the factors under analysis contribute to the environmental deterioration in the Malaysian economy. Therefore, the government may design an environmental or pollution tax policy to minimize carbon emissions. Green energy should be produced and consumed as it is environmental friendly and helps to reduce temperature and atmospheric pollution.

Video Region Annotation with Sparse Bounding Boxes

Video analysis has been moving towards more detailed interpretation (e.g., segmentation) with encouraging progress. These tasks, however, increasingly rely on densely annotated training data both in space and time. Since such annotation is labor-intensive, few densely annotated video data with detailed region boundaries exist. This work aims to resolve this dilemma by learning to automatically generate region boundaries for all frames of a video from sparsely annotated bounding boxes of target regions. We achieve this with a Volumetric Graph Convolutional Network (VGCN), which learns to iteratively find keypoints on the region boundaries using the spatio-temporal volume of surrounding appearance and motion. We show that the global optimization of VGCN leads to more accurate annotation that generalizes better. Experimental results using three latest datasets (two real and one synthetic), including ablation studies, demonstrate the effectiveness and superiority of our method.

Genomics-informed outbreak investigations of SARS-CoV-2 using civet.

The scale of data produced during the SARS-CoV-2 pandemic has been unprecedented, with more than 13 million sequences shared publicly at the time of writing. This wealth of sequence data provides important context for interpreting local outbreaks. However, placing sequences of interest into national and international context is difficult given the size of the global dataset. Often outbreak investigations and genomic surveillance efforts require running similar analyses again and again on the latest dataset and producing reports. We developed civet (cluster investigation and virus epidemiology tool) to aid these routine analyses and facilitate virus outbreak investigation and surveillance. Civet can place sequences of interest in the local context of background diversity, resolving the query into different 'catchments' and presenting the phylogenetic results alongside metadata in an interactive, distributable report. Civet can be used on a fine scale for clinical outbreak investigation, for local surveillance and cluster discovery, and to routinely summarise the virus diversity circulating on a national level. Civet reports have helped researchers and public health bodies feedback genomic information in the appropriate context within a timeframe that is useful for public health.

A study of feature selection methods for android malware detection

Feature Selection (FS) provides a vital role in the android malware detection system. The researchers have presented FS methods and tested them on benchmark datasets, including static types of features extracted from applications. This paper studies FS methods used in traditional android malware detection systems. These FS methods are implemented on the benchmark datasets such as Genome project, Google PlayStore, AndroZoo, and Drebin consist of static types of extracted features from applications. These traditional methods are studied and implemented on the latest dataset, such as CIC-MalDroid2020 dataset, which includes the latest types of malware and 470 dynamic types of features. The experimentation is performed on CIC-MalDroid2020 dataset with the Random Forest (RF) classifier using traditional FS methods, and performance is compared with the original feature set. Finally, the investigation with the RF classifier on CIC-MalDroid2020 dataset using the obtained 80 features from 470 original features by the ReliefF method produces higher precision of 97.4647% and a lower FAR of 0.1409 for malware detection.

Effects of the COVID-19 Pandemic on Dietary Practices for Mothers in Japan

To the best of our knowledge, changes in mothers’ dietary behaviors or perceptions of dietary intake during the COVID-19 pandemic have not been extensively discussed. Therefore, the present study aimed to clarify how the dietary behaviors of mothers living with their children changed during the pandemic in Japan. Ordered probit and interval regressions were applied to individual data for mothers living with children under 18 years of age from five independently conducted Surveys of Attitudes toward Food and Nutrition Education (Shokuiku in Japanese). The estimation results showed that although mothers were more likely to eat dinner with their families during the pandemic, attention to healthy dietary practices, frequency of eating nutritionally balanced food, and dietary practices for preventing lifestyle-related diseases were negatively affected by the pandemic. In conclusion, mothers were more likely to decrease their commitment to a healthy and well-balanced diet during the pandemic than before it. Further research based on the latest datasets with detailed information over the long-term is required.

BEFP: An Extension Recognition System Based on Behavioral and Environmental Fingerprinting

Browser extensions are third-party applications that can customize the browsing experience. Previous studies have shown that browser extension fingerprinting can be used to track users and reveal users’ privacy information by obtaining the browser extension list. However, the proposal of various defense measures weakens the effectiveness of the existing extension fingerprinting technologies. In this paper, we first propose two extension fingerprinting technologies: JavaScript-based environmental fingerprinting and DOM-based behavioral fingerprinting. They, respectively, capture the operation behaviors of extensions on JavaScript properties and webpage’s DOM. Second, we design BEFP, an extension recognition system which comprehensively utilizes the above two technologies to improve the uniqueness of the extension fingerprint. Finally, we collect the latest data set and carry out experiments on the actual scenario where users install multiple extensions. The results show that the true positive rate of extension recognition is as high as 96.3%. And the extension’s detectable rate of BEFP is superior to the existing technologies. Moreover, it is proved that the JavaScript-based environmental fingerprinting can complement the DOM-based fingerprinting to distinguish the extensions with the same DOM modification.

Detection of Network Attacks using Machine Learning: A New Approach

Abstract: The Cyber-attacks become the most important security problems in the today’s world. With the increase in use of computing resources connected to the Internet like computers, mobiles, sensors, IoTs in networks, Big Data, Web Applications/Server, Clouds and other computing resources, hackers and malicious users are planning new ways of network intrusions. Many techniques have been developed to detect these intrusions which are based on data mining and machine learning methods. These intrusions detection techniques have been applied on various IDS datasets. UNSW-NB15 is the latest dataset. This data set contains different modern attack types and wide varieties of real normal activities. In this paper, we compare Naïve Bays algorithm with proposed probability based supervised machine learning algorithms using reduced UNSW NB15 dataset. Keywords: UNSW NB-15, Machine Learning, Naïve Bayes, All to Single (AS) features probability Algorithm

The effects of bank competition, financial stability and ownership structure: evidence from the Middle East and North African (MENA) countries

PurposeThis study aims to show the relationship between competition, financial stability and ownership structure of banks in the Middle East and North African (MENA) countries.Design/methodology/approachThis study uses the generalized method of moments (GMM) estimators to generate research results. This study uses an unbalanced panel dynamic data set. It covers the period 2011 to 2017 in MENA banks.FindingsThis study implies that there is a significant and positive relationship between market power and the financial stability of banks in MENA countries. It explains a competitive market focus on credit risk, which turns them risky. From the bank’s ownership view, Islamic banks are in a less risky position which means Islamic banks are more stable than other ownership structures. On the other hand, government specialized institute displays their poor financial stability and risky from other ownership structures. Unfortunately, there is no significant impact of ownership structure on competition unless Islamic banks prove that they (Islamic banks) perform better in market power.Practical implicationsThe empirical findings of this study suggest that MENA banks should improve the process of managing and monitoring the non-performing loan (loan segment business). It reduces the level of credit risk, which leads to achieving more profit. It also recommends that loan quality should improve immediately in this region for declining financial disruption. Based on the ownership structure, policymakers and stakeholders should adjust their risk and financial stability. Notably, the stakeholders can focus on Islamic banks in this region as this type of ownership structure showing superiority over other ownership structures.Originality/valueThis study is based on the latest data set and produced outcomes by using a GMM estimator. It also uses multiple measures of competition and risk variables to get robust results. Moreover, to the best of the knowledge, this study is the pioneer to examine the competition, risk (financial stability) and ownership structure of banks in the MENA countries.