Search Results (705)

Biomass burning processes affect many semi-rural areas in the Mediterranean, but there is a lack of long-term datasets focusing on their classification, obtained by monitoring carbonaceous particle concentrations and optical properties variations. To address this issue, a campaign to measure equivalent black carbon (eBC) and particle number size distributions (0.3–10 μm) was carried out from August 2019 to November 2020 at a coastal semi-rural site in the Basilicata region of Southern Italy. Long-term datasets were useful for aerosol characterization, helping to clearly identify traffic as a constant eBC source. For a shorter period, PM_2.5 mass concentrations were also measured, allowing the estimation of elemental and organic carbon (EC and OC), and chemical and SEM (scanning electron microscope) analysis of aerosols collected on filters. This multi-instrumental approach enabled the discrimination among different biomass burning (BB) processes, and the analysis of three case studies related to domestic heating, regional smoke plume transport, and a local smoldering process. The AAE (Ångström absorption exponent) daily pattern was characterized as having a peak late in the morning and mean hourly values that were always higher than 1.3. Full article

In today’s fast-paced world of software development, it is essential to ensure that programs run smoothly without any issues. When dealing with complex applications, the objective is to predict and resolve problems before they escalate. The prediction of software defects is a crucial element in maintaining the stability and reliability of software systems. This research addresses this need by combining advanced techniques (ensemble techniques) with seventeen machine learning algorithms for predicting software defects, categorised into three types: semi-supervised, self-supervised, and supervised. In supervised learning, we mainly experimented with several algorithms, including random forest, k-nearest neighbors, support vector machines, logistic regression, gradient boosting, AdaBoost classifier, quadratic discriminant analysis, Gaussian training, decision tree, passive aggressive, and ridge classifier. In semi-supervised learning, we tested are autoencoders, semi-supervised support vector machines, and generative adversarial networks. For self-supervised learning, we utilized are autoencoder, simple framework for contrastive learning of representations, and bootstrap your own latent. After comparing the performance of each machine learning algorithm, we identified the most effective one. Among these, the gradient boosting AdaBoost classifier demonstrated superior performance based on an accuracy of 90%, closely followed by the AdaBoost classifier at 89%. Finally, we applied ensemble methods to predict software defects, leveraging the collective strengths of these diverse approaches. This enables software developers to significantly enhance defect prediction accuracy, thereby improving overall system robustness and reliability. Full article

Nowadays, food traceability represents an important issue in the current context of trade agreements, which influence global food prices. Many consumers prefer to pay a higher price for a traditional cultivation regime of a certain food product that comes from a certain region, appreciating the taste of the respective foodstuff. The potato is now the world’s fourth most important food crop in terms of human consumption, after wheat, maize, and rice. In this context, 100 potato samples from the Romanian market were collected. While 68 samples came from Romania, the rest of the 32 were from abroad (Hungary, France, Greece, Italy, Germany, Egypt, and Poland). The countries selected for potato sample analysis are among the main exporters of potatoes to the Romanian market. The samples were investigated by their multi-elemental and isotopic (²H, ¹⁸O and ¹³C) fingerprints, using Inductively Coupled Plasma Mass Spectrometry (ICP-MS) and Isotope Ratio Mass Spectrometry (IRMS). Then, to distinguish the geographical origin, the experimental results were statistically processed using linear discriminant analysis (LDA). The best markers that emphasize Romanian potatoes were identified to be δ¹³C_bulk, δ²H_water, and Sr. Full article

Negative Poisson’s ratio (NPR) reinforcement offers a novel solution to the usual trade-off between strength gains and ductility loss. Incorporating NPR into ultra-high-performance concrete (UHPC) effectively overcomes the ductility limitations of structural elements. However, the interfacial bonding between NPR reinforcement and UHPC is not sufficiently studied, especially its patterns and mechanisms, impeding the application of the materials. In this paper, the effects of nine design parameters (rebar type, prestrain, etc.) on the bond performance of NPR-UHPC through eccentric pull-out tests are investigated, and a quantitative discriminative indicator K_c for NPR-UHPC bond failure modes is established. The results showed that when K_c ≤ 4.3, 4.3 < K_c ≤ 5.64, and K_c ≥ 5.6, the NPR-UHPC specimens undergo splitting failure, splitting–pull-out failure, and pull-out failure, respectively. In terms of bonding with UHPC, the NPR bars outperform the HRB400 bars, and the HRB400 bars outperform the helical grooved (HG) bars. For the NPR bars, prestrain levels of 5.5%, 9.5%, and 22.0% decrease τ_u by 5.07%, 7.79%, and 17.01% and s_u by 7.00%, 15.88%, and 30.54%, respectively. Bond performance deteriorated with increasing rib spacing and decreasing rib height. Based on the test results, an artificial neural network (ANN) model is developed to accurately predict the critical embedded length l_cd and ultimate embedded length l_ud between NPR bars and UHPC. Moreover, the MAPE of the ANN model is only 53.9% of that of the regression model, while the RMSE is just 62.0%. Full article

The El Bola mafic–ultramafic intrusion (EBMU) in Egypt’s Northern Eastern Desert represents an example of Neoproterozoic post-collisional layered mafic–ultramafic magmatism in the Arabian–Nubian Shield (ANS). The intrusion is composed of pyroxenite, olivine gabbro, pyroxene gabbro, pyroxene–hornblende gabbro, and hornblende-gabbro, exhibiting adcumulate to heter-adcumulate textures. Mineralogical and geochemical analyses reveal a coherent trend of fractional crystallization. Compositions of whole rock and minerals indicate a parental magma of ferropicritic affinity, derived from partial melting of a hydrous, metasomatized spinel-bearing mantle source, likely modified by subduction-related fluids. Geothermobarometric calculations yield crystallization temperatures from ~1120 °C to ~800 °C and pressures from ~5.2 to ~3.1 kbar, while oxygen fugacity estimates suggest progressive oxidation (log fO₂ from −17.3 to −15.7) during differentiation. The EBMU displays Light Rare Earth element (LREE) enrichment, trace element patterns marked by Large Ion Lithophile Element (LILE) enrichment, Nb-Ta depletion and high LILE/HFSE (High Field Strength Elements) ratios, suggesting a mantle-derived source that remained largely unaffected by crustal contribution and was metasomatized by slab-derived fluids. Tectonic discrimination modeling suggests that EBMU magmatism was triggered by asthenospheric upwelling and slab break-off. Considering these findings alongside regional geologic features, we propose that the mafic–ultramafic intrusion from the ANS originated in a tectonic transition between subduction and collision (slab break-off) following the assembly of Gondwana. Full article

Ensuring cybersecurity and health management is a fundamental requirement in modern chemical industry plants operating under the Industry 4.0 framework. Traditionally, these two concerns have been addressed independently, despite sharing multiple underlying elements which suggest the viability of a unified detection and localization solution. This study introduces a computational intelligence framework based on fuzzy techniques, which allows for the early identification and precise localization of both faults and cyberattacks, along with the capability to recognize previously unseen events during runtime. Once new events are identified and classified, the training database is updated, creating a mechanism for continuous learning. This integrated approach simplifies the computational complexity of supervisory systems and enhances collaboration between the Operational Technology and Information Technology teams within chemical plants. The proposed methodology demonstrates strong robustness and reliability, even in complex conditions characterized by noisy measurements and disturbances, achieving outstanding performance due to its excellent discrimination capabilities. Full article

Infrared small target detection (IRSTD) aims to locate and separate targets from complex backgrounds. The challenges in IRSTD primarily come from extremely sparse target features and strong background clutter interference. However, existing methods typically perform discrimination directly on the features extracted by deep networks, neglecting the distinct characteristics of weak and small targets in the frequency domain, thereby limiting the improvement of detection capability. In this paper, we propose a frequency-aware masked-attention network (FM-Net) that leverages multi-scale frequency clues to assist in representing global context and suppressing noise interference. Specifically, we design the wavelet residual block (WRB) to extract multi-scale spatial and frequency features, which introduces a wavelet pyramid as the intermediate layer of the residual block. Then, to perceive global information on the long-range skip connections, a frequency-modulation masked-attention module (FMM) is used to interact with multi-layer features from the encoder. FMM contains two crucial elements: (a) a mask attention (MA) mechanism for injecting broad contextual feature efficiently to promote full-level semantic correlation and focus on salient regions, and (b) a channel-wise frequency modulation module (CFM) for enhancing the most informative frequency components and suppressing useless ones. Extensive experiments on three benchmark datasets (e.g., SIRST, NUDT-SIRST, IRSTD-1k) demonstrate that FM-Net achieves superior detection performance. Full article

The authentication of high-value spices such as paprika and cinnamon is critical due to increasing food fraud. This study explored the potential of a multi-analytical approach, combined with chemometric tools, to differentiate 45 paprika and 46 cinnamon samples from the Slovenian market based on their geographic origin, production methods, and possible adulteration. The applied techniques included stable isotope ratio analysis (δ¹³C, δ¹⁵N, δ³⁴S), multi-elemental profiling, FTIR, and antioxidant compound analysis. Distinct isotopic and elemental markers (e.g., δ¹³C, δ³⁴S, Rb, Cs, V, Fe, Al) contributed to classification by geographic origin, with preliminary classification accuracies of 90% for paprika (Hungary, Serbia, Spain) and 89% for cinnamon (Sri Lanka, Madagascar, Indonesia). Organic paprika samples showed higher values of δ¹⁵N, δ³⁴S, and Zn, whereas conventional ones had more Na, Al, V, and Cr. For cinnamon, a 95% discrimination accuracy was achieved between production practice using δ³⁴S and Ba, as well as As, Rb, Na, δ¹³C, S, Mg, Fe, V, Al, and Cu. FTIR differentiated Ceylon from cassia cinnamon and suggested possible paprika adulteration, as indicated by spectral features consistent with oleoresin removal or azo dye addition, although further verification is required. Antioxidant profiling supported quality assessment, although the high antioxidant activity in cassia cinnamon may reflect non-phenolic contributors. Overall, the results demonstrate the promising potential of the applied analytical techniques to support spice authentication. However, further studies on larger, more balanced datasets are essential to validate and generalize these findings. Full article

Late Paleozoic to Early Mesozoic granitoids in the East Kunlun Orogenic Belt (EKOB) provide critical insights into the complex and debated relationship between Paleo–Tethyan magmatism and tectonics. This study presents integrated bulk-rock geochemical and zircon isotopic data for the Xingshugou monzogranite (MG) to address these controversies. LA-ICP-MS zircon U-Pb dating constrains the emplacement age of the MG to 247.1 ± 1.5 Ma. The MG exhibits a peraluminous and low Na₂O A₂-type granite affinity, characterized by high K₂O (4.69–6.80 wt.%) and Zr + Nb + Ce + Y (>350 ppm) concentrations, coupled with high Y/Nb (>1.2) and A/CNK ratios (1.54–2.46). It also displays low FeO^T, MnO, TiO₂, P₂O₅, and Mg^# values (26–49), alongside pronounced negative Eu anomalies (Eu/Eu* = 0.37–0.49) and moderately fractionated rare earth element (REE) patterns ((La/Yb)_N = 3.30–5.11). The MG exhibits enrichment in light rare earth elements (LREEs) and large ion lithophile elements (LILEs; such as Sr and Ba), and depletion in high field strength elements (HFSEs; such as Nb, Ta, and Ti), collectively indicating an arc magmatic affinity. Zircon saturation temperatures (T_Zr = 868–934 °C) and geochemical discriminators suggest that the MG was generated under high-temperature, low-pressure, relatively dry conditions. Combined with positive zircon ε_Hf(t) (1.8 to 4.7) values, it is suggested that the MG was derived from partial melting of juvenile crust. Synthesizing regional data, this study suggests that the Xingshugou MG was formed in an extensional tectonic setting triggered by slab rollback of the Paleo-Tethys Oceanic slab. Full article

The project of the flexible direct transmission of renewable energy has become an inevitable development trend for the large-scale grid connection of renewable energy. Its two-terminal weakly fed AC system is often composed of 100% power electronic equipment, which leads to an essential transformation in fault characteristics and protection requirements. At present, in research, the traditional directional elements are limited by the negative-sequence control strategy, resulting in the decline of their sensitivity and reliability. Therefore, this paper proposes a model for identifying directional elements using composite electrical quantities that is not affected by the control strategy of the two-terminal weakly fed AC system and can reliably identify the fault direction. Firstly, the adaptability of traditional directional elements under the negative-sequence current suppression strategy on both sides of the system when faults occur in the AC line was analyzed. Secondly, based on the idea of model recognition, the model relationship of fault voltage and current in the case of ground faults and non-ground faults occurring at different locations was analyzed. Finally, a fitted voltage was constructed and the Kendall correlation coefficient was introduced to achieve fault direction discrimination. Simulation results demonstrate that the proposed pilot protection scheme can operate reliably under conditions of 300 Ω transition resistance and 25 dB noise interference. Full article

Global climate change is reshaping the habitat suitability of medicinal plants, potentially compromising their phytochemical integrity and therapeutic efficacy. Siraitia grosvenorii, an edible medicinal plant in China, has expanded its cultivation area into non-native habitats. Therefore, this study analyzed the suitable cultivation region under different periods in China based on the MaxEnt model, and 59 samples were investigated to explore the interrelationships between chemical constituents and climatic variables through multivariate statistical analysis, which will contribute to meeting the sustainable supply of high-quality S. grosvenorii. We discovered that appropriate habitats cover an area of 58.76 × 10⁴ km², mainly in the southern parts of China. Under future climate conditions, suitable habitats decrease and shift to the northeast along the current habitats. The precipitation levels of the driest month, precipitation seasonality, and temperature seasonality were crucial for its distribution. Furthermore, 11 elements were identified to distinguish samples from different suitable areas through orthogonal partial least squares discriminant analysis. Correlation analysis revealed a strong association between chemical constituents and various climatic factors. This study offers valuable insights into potential S. grosvenorii cultivation areas in China and provides reference indicators for quality evaluation. Full article

The Belle II experiment has collected a 364 fb⁻¹ sample of collisions at the ${\rm Y}\left(4S\right)$ resonance. This dataset, with its low particle multiplicity and well-constrained initial state, provides an ideal environment for studying semileptonic and missing energy B decays. In this paper, I will present recent results on these decays, emphasizing their impact on the determination of CKM matrix elements and potential new physics. I will also discuss the techniques used for missing energy reconstruction and the challenges of signal-background discrimination. Future analysis prospects with larger datasets will also be highlighted. Full article

This study presents an innovative approach to differentiate Southern Greek yellow prickly pear samples according to geographical origin based on physicochemical parameters, mineral composition, and bioactive compounds using advanced chemometrics. A total of 56 yellow prickly pear samples were collected from four distinct Greek regions (Crete, Paros, Symi, Peloponnese) during the 2019 and 2020 harvest seasons. A multi-platform analytical strategy was employed, combining classical physicochemical analyses and UV spectrophotometry for total antioxidant activity with cutting-edge techniques such as UPLC-MS/MS for precise quantification of vitamins and antioxidants, and ICP-MS for mineral profiling. In total, thirteen physicochemical parameters, nineteen macro-, micro-, and trace elements, nine vitamins, and seven antioxidants were identified and quantified. Application of MANOVA and Linear discriminant analysis (LDA) revealed that eight physicochemical parameters, ten mineral elements, and sixteen bioactive compounds played a crucial role in sample geographical differentiation. The classification success rates using the cross-validation method were 82.1% for physicochemical parameters, 75.0% for minerals, and an impressive 96.4% for vitamins and antioxidants highlighting the robust tool for the geographical differentiation of Southern Greek yellow prickly pears. Full article

The widespread scrapping of retired mechanical parts has led to severe waste of resources and environmental burdens, posing a significant challenge to sustainable industrial development. To enable efficient recycling of retired mechanical parts and enhance the sustainability of their remanufacturing processes, the concept of biological genes is adopted to characterize the changes in the information of retired mechanical parts during the remanufacturing process as gene mutations of parts, aiming to maximize remanufacturing potential and devise an optimal generalized remanufacturing strategy for extending part life cycles. However, gene mutation of retired mechanical parts is not an isolated event. The modification of local genes may disrupt the original equilibrium of the part’s state, leading to conflicts such as material–performance, structure–function/performance, and function–performance. These conflicts constitute a major challenge and bottleneck in designing generalized remanufacturing schemes. Therefore, we propose a conflict identification and resolution method for gene mutation of retired mechanical parts. First, gene mutation graph of retired mechanical parts is established to express its all-potential remanufacturing pathways. Using discrimination rules and the element representation method from extenics, mutation conflicts are identified, and a conflict problem model is constructed. Then, the theory of inventive problem solving (TRIZ) engineering parameters are reconstructed and mapped to the mutation conflict parameters. The semantic mapping between the inventive principles and the transforming bridges is established by the Word2Vec algorithm, thereby improving the transforming bridge method to generate conflict resolution solutions. A coexistence degree function of transforming bridges is proposed to verify the feasibility of the resolution solutions. Finally, taking the generalized remanufacturing of a retired gear shaft as an example, we analyze and discuss the process and outcome of resolving gene mutation conflicts, thereby verifying the feasibility and effectiveness of the proposed concepts and methodology. Full article

In this paper, we present a stance detection model grounded in multi-task learning, specifically designed to address the intricate challenge of text stance analysis within social media comments. This model is structured with an embedding network, an encoder module, a sophisticated multi-task attention mechanism, an ensemble module, and a classification output layer. To augment the performance of stance detection, we employed sentiment analysis and toxicity language detection as auxiliary tasks. The sentiment analysis plays a pivotal role in enabling the model to capture the public opinion inclinations of both individual and collective users. By delving into these inclinations, our model can extract fine-grained stance elements, offering a more nuanced understanding of users’ positions. On the other hand, toxicity language detection aids in modeling the extreme tendencies of social media users towards specific events. It identifies manifestations of hatred, offensiveness, discrimination, and insult, thereby allowing the model to reconstruct users’ genuine stance information from these extreme expressions. Through the synergy of multi-task joint learning, the accuracy and reliability of the stance detection were significantly improved. To validate the efficacy of our proposed model, we selected two hot events as representative cases, one from the Chinese Weibo platform and the other from the English Twitter platform. A series of comprehensive tasks, including developing crawler programs, collecting data, performing data preprocessing, and conducting data annotation, were systematically executed. Subsequently, we applied our model to detect the stances within the comments related to these two events, categorizing them into three classes: support, opposition, and ambiguity. The experimental results demonstrate that our stance detection model, which integrates sentiment analysis and toxicity language detection, substantially improves the detection accuracy, outperforming traditional methods. Full article