RV Coefficient Research Articles

Environmental DNA Metabarcoding Detects Predators at Higher Rates Than Electrofishing

ABSTRACTThere are numerous downsides and risks associated with electrofishing; hence, environmental DNA (eDNA) metabarcoding is becoming increasingly common in aquatic ecological studies. Generally, researchers agree that eDNA metabarcoding is more sensitive than electrofishing, and that eDNA metabarcoding is better at detecting rare species. As predatory species tend to be rarer than prey species, eDNA metabarcoding should hypothetically detect more predator species than electrofishing. Instead of supporting the notion that eDNA must replace electrofishing, or that eDNA and electrofishing must display the same results, the current study aims to establish the strengths and weaknesses of eDNA metabarcoding when compared to electrofishing. eDNA metabarcoding and electrofishing data were collected on three sampling dates at four experimental sites. A RV coefficient analysis confirmed that the eDNA metabarcoding data (RV = 0.395, p = 0.057) are statistically different from the electrofishing data. A paired Wilcoxon signed rank test revealed that eDNA data collection techniques detect more predatory species than electrofishing (p = 0.041). When the analysis was conducted for prey species a statistically significant difference did not occur (p = 0.661). Overall, the results of the study suggest that eDNA metabarcoding does not display the same results as electrofishing due to eDNA metabarcoding detecting predatory species at higher rates. The combined use of eDNA alongside electrofishing can help mitigate electrofishing's bias against predatory species, while electrofishing can address reliability concerns associated with eDNA. This collaborative approach ultimately enhances the accuracy of fish community assessments.

Temporal dominance of Sensations: Do different concepts of “dominance” affect the results?

The purpose of this study was to understand the role of “dominance” definitions in the results of TDS applied to consumers. This study compared two temporal dominance of sensations (TDS) approaches − TDS-I − that attracts the most attention and TDS-II − most intense/strongest sensation − in the context of evaluating artisanal Minas cheeses from the Serra da Canastra and chocolate with different cocoa concentrations samples. TDS curves were constructed and a trajectory-based principal component analysis (PCA) was performed dominance rates at ten equally spaced time points. Additionally, difference curves, multiple factor analysis (MFA) and regression vector coefficient (RV coefficient) were performed to compare the two approaches. The findings showed that the two approaches produced similar results, suggesting that consumers interpret the terms dominant and intense in a similar way during TDS evaluations, the results were even closer at the chocolate study than at the cheese study. However, in both approaches were observed low dominance rates and differences between the sensation perception time, mainly in the evaluation of cheese flavor. This variability may be attributed to the complexity of the cheese’s flavor and the varied interpretations of dominance among evaluators. Despite these differences, the approaches showed similar characterizations across the same samples, demonstrating high reproducibility and a strong ability to differentiate between samples. This study demonstrates that the choice between the terms dominant (which captures more attention) or intense (stronger) to guide consumers in performing the sensory test does not significantly influence the results. Consequently, it is possible to adopt greater freedom and flexibility in the terminology used to instruct participants in conducting the test.

Sella turcica and facial bones: Morphological integration in the human fetal cranium.

The cranial base plays a significant role in facial growth, and closer analyses of the morphological relationship between these two regions are needed to understand the morphogenesis of the face. Here, we aimed to study morphological integration between the sella turcica (ST) and facial bones during the fetal period using geometric morphometrics. Magnetic resonance images of 47 human fetuses in the Kyoto Collection, with crown-rump lengths of 29.8-225 mm, were included in this study. Anatomical homologous landmarks and semilandmarks were registered on the facial bones and the midsagittal contour of the ST, respectively. The shape variations in the craniofacial skeleton and the ST were statistically investigated by reducing dimensionality using principal component analysis (PCA). Subsequently, the morphological integration between the facial bones and ST was investigated using two-block partial least squares (2B-PLS) analysis. PCA showed that small specimens represented the concave facial profile, including the mandibular protrusion and maxillary retrusion. The 2B-PLS showed a strong integration (RV coefficient = 0.523, r = .79, p < .01) between the facial bones and ST. The curvature of the anterior wall of the ST was highly associated with immature facial morphology characterized by a concave profile. The strong integration between the two regions suggested that the anterior ST may be associated with facial morphology. This result quantitatively confirms previous studies reporting ST deformities in facial anomalies and induces further research using postnatal subjects.

Exploring food and beverage pairing from a cross-cultural projective mapping

Culture is a well-known driver of food choices, and therefore, it could also impact food pairing preferences. Food pairing has been studied from different approaches; however, little cross-cultural research has been done. This work explored food and beverage pairing using projective mapping (PM) to create maps of food-beverage combinations. Four countries (Mexico, Argentina, France, and Norway), thirty foods, and six beverages were selected. PM was carried out through an online study in each country. Participants were asked to map foods together with beverages following the instruction that foods and beverages closer together represented a good combination. The coordinates of each product were analyzed through Multiple Factorial Analyses (MFA) by countries. The first four factors of each MFA were used to perform RV coefficients to test similarities in food-beverage pairings between the countries. Finally, a k-means clustering was performed on the beverage coordinates of each MFA. PM provided maps representing food and beverage pairings for each country in which the proximity between food-beverages represented a good combination according to consumers. RV coefficients between countries were low, showing that food-beverage pairings were not similar across countries, evidencing the cultural effect in food-drink combinations. Results from the k-means clustering showed some similarities and differences between countries. In general, the food-beverage pairing was effectively explored with PM, from which several differences and similarities were found within cultures.

Online consumer testing beyond central location tests: A case study for brewed coffee

Due to the constraints of the COVID-19 pandemic, conducting sensory evaluations requiring direct interactions became challenging. In response, researchers have been motivated to devise non-face-to-face testing methods as alternatives. This study aimed to compare two non-face-to-face home-use tests (HUT) with the traditional face-to-face central location test (CLT). Both HUTs involved online recruitment and sample delivery to participants' homes. One HUT provided a written protocol with no direct interaction (contactless HUT; C-HUT), whereas the other included an online meeting with a researcher for live guidance (online HUT; O-HUT). Four coffee samples were evaluated on the basis of liking and sensory and emotional attributes. The comparison between CLT and O-HUT showed RV coefficients of 0.92, 0.93, and 0.98 (P < 0.05) for liking and sensory and emotional attributes, respectively. In addition, based on the RV coefficient, the CLT results showed a significantly greater similarity to those of O-HUT compared to those of C-HUT. The O-HUT also outperformed the C-HUT in its ability to significantly discriminate between samples. Hence, real-time interactions between researchers and participants, as facilitated by O-HUT, may be more suitable in certain scenarios compared to C-HUT, which relies solely on a written protocol. Overall, these findings suggest that C-HUT and O-HUT are suitable methods for collecting sensory data and overcoming geographic and face-to-face contact limitations, providing greater flexibility, and reducing the time and cost associated with traditional sensory evaluations.

Microbial landscape of cooked meat products: evaluating quality and safety in vacuum-packaged sausages using culture-dependent and culture-independent methods over 1 year in a sustainable food chain.

Over the last few decades, advancements in process safety and quality methods have been significantly improved, yet new challenges continue to emerge in the sustainable food supply chain. This study aimed to investigate some physicochemical and microbiological parameters impacting meat products, particularly cooked sausages, within a sustainable supply chain, focusing on quality, spoilage populations, and syneresis formation under vacuum conditions. A comprehensive analysis was conducted on 355 samples collected over four seasons using high-throughput sequencing (16S/ITS) and microbiological and physicochemical [pH and water activity (aw)] assessments. The microbial growth predictor MicroLab_ShelfLife was employed, and multiple factor analysis (MFA) and agglomerative hierarchical clustering (AHC) were utilized to understand how these variables influence the microbiome resilience of these products. Lactic and acetic acids were correlated with the microbiome of the sausages and the liquid coating covering them using metagenomic analyses. The study highlighted that 52% of the evaluated meat industries in southeastern Brazil are implementing effective protocols for sustainable chain production. The results indicated that the durability of vacuum-packaged cooked sausages was primarily influenced by storage temperature (RV coefficient of 0.906), initial microbial load (0.755), and aw (0.624). Average microbial counts were 4.30 log cfu/g (initial), 4.61 (7°C/4 days), 4.90 (7°C/8 days), 6.06 (36°C/4 days), and 6.79 (36°C/8 days). Seasonal durability analysis revealed that winter had the highest average durability of 45.58 days, while summer had the lowest at 26.33 days. Yeast populations, including Trichosporon sp. and Candida sp., were identified as key genera influencing spoilage dynamics. In addition, Bacillus species emerged as dominant spoilage microorganisms, highlighting the need for new critical controls. This study demonstrates the impact of metagenomic approaches, including ITS and 16S amplicon sequencing, in revealing microbial community dynamics, storage temperature, and aw, which are essential for developing targeted interventions to enhance food safety and quality sustainably.

Carbon and Nutrients from Organic Residues Modulate the Dynamics of Prokaryotic and Fungal Communities.

Inputs of carbon (C) and nutrients from organic residues may select specific microbes and shape the soil microbial community. However, little is known about the abiotic filtering of the same residues with different nutrient concentrations applied to the soil. In our study, we explored how applying organic residue, vinasse, as fertilizer in its natural state (V) versus its concentrated form (CV) impacts soil microbiota. We conducted two field experiments, evaluating soil prokaryotic and fungal communities over 24 and 45 days with vinasse (V or CV) plus N fertilizer. We used 16S rRNA gene and ITS amplicon sequencing. Inorganic N had no significant impact on bacterial and fungal diversity compared to the control. However, the varying concentrations of organic C and nutrients in vinasse significantly influenced the soil microbiome structure, with smaller effects observed for V compared to CV. Prokaryotic and fungal communities were not correlated (co-inertia: RV coefficient = 0.1517, p = 0.9708). Vinasse did not change the total bacterial but increased the total fungal abundance. A higher C input enhanced the prokaryotic but reduced the fungal diversity. Our findings highlight vinasse's role as an abiotic filter shaping soil microbial communities, with distinct effects on prokaryotic and fungal communities. Vinasse primarily selects fast-growing microorganisms, shedding light on the intricate dynamics between organic residues, nutrient concentrations, and soil microbes.

Congruence of water column, contemporary and pre‐industrial sediment cladoceran assemblages in 85 Canadian lakes of contrasting human impact levels

Abstract Crustacean zooplankton occupy a key position as primary consumers within lake food webs, acting as early indicators of wider ecosystem shifts that can cascade to higher trophic levels. Their communities can be assessed via water column net hauls or via sediment samples containing preserved subfossil remnants from various time periods; the latter probably integrate multiple years of data, habitats within lakes, and seasonal patterns. The degree to which such samples do or do not correspond has never been assessed across a wide variety of lake types and gradients. As part of a larger sampling of over 600 lakes across lake‐rich Canada, 85 lakes were selected along a gradient of human impact, according to contemporary catchment land use, to analyse cladoceran zooplankton assemblage relatedness among snapshot water column, contemporary and pre‐industrial sediment samples. We used a combination of ordinations and matrix correlations (RV coefficients), along with partial redundancy analyses, to determine the degree to which the different sample types were comparable and whether they responded similarly to &gt;30 environmental variables. We also compared conclusions obtained using the full cladoceran assemblages versus pelagic‐only taxa and whether our results varied according to degree of human impact. Overall, while all sample types were correlated, congruence was highest between contemporary and pre‐industrial sediment sample types. This result could arise from the under‐representation of littoral taxa in water column samples relative to those from the sediment. In high impacted catchments, however, we observed greater similarity between water column and contemporary sediment samples, as water quality stressors potentially structure assemblages in parallel directions over time. In these modern‐day sample types and among measured variables, we identified lake depth and pH as the most important structuring variables across Canada. In this first study comparing cladoceran assemblages in water column, contemporary and pre‐industrial sediments across many lakes, we conclude that water column and sediment samples respond similarly to environmental drivers but should not be considered entirely analogous or redundant.

Exact first moments of the RV coefficient by invariant orthogonal integration

The RV coefficient measures the similarity between two multivariate configurations, and its significance testing has attracted various proposals in the last decades. We present a new approach, the invariant orthogonal integration, permitting to obtain the exact first four moments of the RV coefficient under the null hypothesis.Our proposal can be applied to any multivariate setting endowed with Euclidean distances between the observations. It also covers the weighted setting of observations of unequal importance, where the exchangeability assumption, justifying the usual permutation tests, breaks down.The proposed RV moments express as simple functions of the kernel eigenvalues occurring in the weighted multidimensional scaling of the two configurations (spectral effective dimensionality, spectral skewness and spectral excess kurtosis). The expressions for the third and fourth moments seem original, and explain the marked asymmetry and kurtosis of the RV coefficient. They permit to test the significance of the RV coefficient by Cornish–Fisher cumulant expansion, beyond the normal approximation, as illustrated on a small dataset.The first three moments can be obtained by elementary means, but computing the fourth moment requires a more sophisticated apparatus, the Weingarten calculus for orthogonal groups.

SVM-RFE enabled feature selection with DMN based centroid update model for incremental data clustering using COVID-19

This research introduces an efficacious model for incremental data clustering using Entropy weighted-Gradient Namib Beetle Mayfly Algorithm (NBMA). Here, feature selection is done based upon support vector machine recursive feature elimination (SVM-RFE), where the weight parameter is optimally fine-tuned using NBMA. After that, clustering is carried out utilizing entropy weighted power k-means clustering algorithm and weight is updated employing designed Gradient NBMA. Finally, incremental data clustering takes place in which centroid matching is carried out based on RV coefficient, whereas centroid is updated based on deep maxout network (DMN). Also, the result shows the better performance of the proposed method..

Comparison of descriptive analysis and flash profile by naïve consumers and experts on commercial milk and yogurt products

Flash profile (FP), a ranking-based consumer-oriented method, has been studied as a rapid consumer profiling method that can be used as an alternative to conventional descriptive analyses, which require significant time, cost, and effort to get results. However, few studies have compared conventional profiling methods for different FP panel groups. In this study, we investigated the similarities between three methodologies, conventional descriptive analysis, FP by consumers, and FP by experts, on two commercial products: milk and yogurt. The results showed minor qualitative differences in the sensory attributes generated by the assessors between the three methodologies. The expert groups used terms related to product-specific attributes, whereas consumers used less specific terms. However, the overall results of the sample configurations and RV coefficients of the three methodologies were similarly high, ranging from 0.92 to 0.97 for both sample sets. These results highlight the applicability of FP when conducted with experts and consumers. This study showed that consumer and expert panels generated comparable sample configuration results using FP compared to the descriptive analysis conducted by trained panelists in two food matrices.

A novel hybrid optimization algorithm for depression detection using MRI and speech signal

Depression is defined as a psychiatric disorder, which has many adverse effects on the person, family, and society. In this paper, the multimodalities are utilized as an input for depression detection that includes both speech signal and brain MRI for better detection. Here, the median filter and Gaussian filter are utilized to remove the clamour exist in MRI and speech signal, respectively. State of depression is classified from both MRI and speech signal using a Deep Maxout Network (DMN), where the weight of the deep learning network is trained by a novel approach of Exponential African Vulture Optimization Algorithm (ExpAVOA). The ExpAVOA is formed by the merging of the Exponentially Weighted Moving Average (EWMA) and African Vultures Optimization Algorithm (AVOA). Finally, the classified outcomes of the MRI and speech signals are fused by the RV coefficient. Thus, the classified outcome is either a healthy person or a depressed person. The unique optimal deep learning technique attained maximum testing accuracy, sensitivity, and specificity of 0.959, 0.942, and 0.968 respectively.

Consumer testing away from a sensory facility: Application of home-use test and no-contact home-use test

COVID-19 pandemic-related restrictions for approximately three years have heavily influenced sensory evaluations. People have become accustomed to working remotely and communicating online. This has led to opportunities in sensory testing paired with logistics systems and information technologies, resulting in a wide application of the home-use test (HUT), wherein panelists evaluate samples from their homes or other off-site locations. This study aimed to compare three sensory evaluation conditions: a central location test (CLT, n = 104), a HUT (n = 120), and a no-contact HUT (N-HUT, n = 111). We recruited participants via the local community website, delivered samples using a delivery service, and conducted sensory testing using a smartphone for the N-HUT. Participants were requested to report the acceptance ratings, sensory profiles, and emotion responses to four coffee samples. Some differences in the acceptance ratings might be due to the different attitudes participating in the evaluation. In the sensory profiling of the samples, multi-factor analysis (MFA) revealed highly similar sensory characteristics across the three types of tests. All RV coefficients (RVs) among the test conditions were above 0.93. The emotion responses to coffee samples were similar among test conditions based on the MFA with RV values greater than 0.84. In conclusion, we found that N-HUT produced similar results regarding the descriptions of sensory profiles and emotions, indicating that N-HUT is a suitable test method for collecting sensory data and overcoming CLT and HUT's regional limitations. Modern logistics systems and information technologies make it possible to conduct nationwide sensory evaluations without in-person contact or participant attendance at sensory testing facilities.

From Fermented Wash to New Make Spirit: Assessing the Evolution of Flavour Characteristics of Scotch Whisky Using Lab-Scale Process Simulations

New product development for distilled spirits frequently involves experimental trials at the laboratory scale that attempt to replicate industrial-scale production processes. This process is time-consuming and limits the number of samples that can be analysed. The aim of the present study was to conduct laboratory-scale Scotch malt whisky production experiments to determine if samples taken from earlier in the production process, that is, directly after fermentation (wash stage) or after a single distillation (low wines stage), showed similar analytical differentiation compared to samples of fresh distillates (new make spirits). Napping, a rapid sensory method, was used to assess the impact on flavour characteristics while solid-phase microextraction (SPME) gas chromatography–mass spectrometry (GC–MS) was used to analyse volatile flavour congeners. Hierarchical Multiple Factor Analysis was used to compare the product maps from samples taken at each whisky production stage and revealed a pattern of differences across the samples that could be tracked through the process. Although the flavour descriptors and volatile congeners composition changed at each stage, there were only marginal changes in the differentiation between samples, resulting in the same sample groups being found in all analyses. RV coefficients &gt;0.90 for all analytical comparisons and &gt;0.74 overall showed that all product maps were highly similar to each other and showed the same overall differentiation between samples. These results indicate that the analysis of fermented malt whisky wash may provide sufficient information to proceed to larger-scale industry trials, saving time and allowing a greater number of parameters to be explored.

Basicranial Modular Organization. A Study in the Araucanian Horse of Colombia.

The skull is divided into neurocranium and splanchnocranium, and its variation allows ecomorphological studies to learn about possible evolutionary and adaptive characteristics. The basicranial organization of the neurocranium and splanchnocranium modules was studied in a sample of 31 skulls from adult Araucanian horses by means of 2D geometric morphometric techniques. The neurocranium and splanchnocranium modules on the ventral aspect were analyzed separately using a set of 31 landmarks. The RV coefficient (the multivariate analog of a correlation) was estimated to analyze the independence of these two parts, as well as their morphological integration, using a two-block analysis of least squares. The study results confirm the modular development of the neurocranium and the splanchnocranium, the former being more stable than the latter as well as low morphological integration between the two. The development between both parties is structured in a modular way but allows relative independence. Now it would be interesting for future studies to add muscles (those that connect the cranial parts, but also the cervical), the hyoid apparatus, and the ossicles of the internal ear and the jaw and analyze if they behave as integrated modules between them. Since this research has been conducted at the subspecific breed level, it could be plausible that in other breeds, this integrative development was different.

Morphological Integration on the Calcaneum of Domestic Sheep (Ovis aries Linnaeus, 1758)—A Geometric Morphometric Study

Morphological integration and modularity refer to the degree of covariation between the different components of an anatomical structure. Modularity refers to structures that have components which covary strongly, but at the same time are relatively independent. Integration refers to the coordinated variation of the components of a functional anatomical structure. The hindlimb basipodium is a complex functional structure. The integration of its parts must arise from a coordinated development and functionality. Our objective in this study was to study the levels of integration of two modules on the calcaneum in domestic sheep. The calcaneum develops from two different centres. One gives rise to the body and its process while the other gives rise to the distal half of the bone. The hypothesis of modularity of two parts of the calcaneum was tested using the Escoufier RV coefficient and an analysis of two blocks of Partial Least Squares. These allowed us to evaluate the level of morphological integration. For this purpose, digital images of the medial aspect of complete calcanea of 47 domestic sheep (Ovis aries) were used. Twenty 2D coordinates of homologous anatomical landmarks (4 for the body and the process and 16 for the distal half of the bone) were collected as morphometric data. These were studied using geometric morphometrics. The results indicated good evidence of modular organization and a medium morphological integration between the two modules. Thus, according to our results, the two modules seem to exist and behave in a rather independent way. They appear to share functions rather than precursors of development. This study is important in order to elucidate the underlying factors in the processes of the development of the sheep calcaneum.

SSLnDO-Based Deep Residual Network and RV-Coefficient Integrated Deep Fuzzy Clustering for Cotton Crop Classification

The excellent classification accuracy of remote sensory images has led to a major expansion of their use in agricultural field surveillance in recent years. The accurate classification of crop is very significant for agricultural management and food security. There are diverse varieties of crops are available. Among them, each possesses same spectral curves. Due to increasing number of remote sensing information, a major limitation arises in the domain of crop classification is how to determine significant data from huge volume of information to maintain categorization accuracy and implementation time. Consequently, one of the main obstacles in the agricultural industry is the precise classification of crops. Deep Residual Network (ResNet)-based Social Sea Lion Driver Optimization (SSLnDO)-based Deep Fuzzy Clustering (DFC), where the distance is calculated using RV coefficient, are used to develop an effective method for classifying cotton crops. Finally, classification of the cotton crop is successfully accomplished using Deep ResNet, and the network classifier is trained by inserting produced SSLnDO. The vegetation index is generated for each color band independently. Combining the Social Ski Driver (SSD) optimization and the Sea Lion Optimization (SLnO) method yields the newly developed SSLnDO. A maximum testing accuracy of 0.940, a sensitivity of 0.931, and a specificity of 0.926 have been achieved by the designed model. When comparing the accuracy value produced by the proposed method for image-2 to the existing approaches, namely WLI-Fuzzy + BS-Lion NN, OSVM-OCNN, GAN, Hybrid CNN-RF, Deep learning, and Deep ResNet, they are 15.21%, 11.06%, 15.21%, 9.04%, 4.36%, and 2.3% higher.

A Preliminary Study on the Validity and Stability of Projective Methods: An Application of the Structural Approach of Social Representations with Traditional Mexican Cheeses.

Projective methods are qualitative tools used to study food consumer behavior. In recent years, there has been an increase in studies that use these tools to analyze consumer behavior, particularly with the word association (WA) technique. However, one of the challenges in using these methods is the stability and validity of the data. This research aimed to obtain preliminary information on the stability and validity of the associations generated by consumers with the WA technique, using the structural approach of social representations. For this, two studies were carried out; for the first study, a face-to-face survey was carried out in which 89 consumers participated, who wrote the first words that came to mind with the stimulus "Aculco" on a ballot paper. For the second study, 122 consumers completed the same task as in the first study; however, the participants were recruited from an online survey. A random sample (n = 50) of both studies was selected to explore the stability and validity of the results. In the three study samples, the words were grouped into categories and analyzed through the structural approach of social representations. The frequency of mention of the identified categories was compared with the chi-square test and the average position (AP) and the Cognitive Salience Index (CSI) were calculated. Prototype maps were built to study the structure of the categories according to the central core and peripheral areas. Cluster analysis was performed to corroborate the structure of the representations. Finally, multiple factor analysis (MFA) was performed to determine the similarity of the results obtained from the three samples using the RV coefficient. No statistical differences (p &gt; 0.05) were identified in most of the representations (n = 11) generated from the WA task. Furthermore, the representation "Cheese and dairy products" was positioned in the central core of the three maps. The APs and the CSIs of each representation were similar in the three study samples. The RV coefficient (≥0.80) indicated similarity in the representations obtained. Results of this research can be useful for future studies that attempt to compare the stability and validity of the information based on qualitative and more flexible methodologies. Some methodological implications related to the validity and stability of projective methods are discussed.

FACVO-DNFN: Deep learning-based feature fusion and Distributed Denial of Service attack detection in cloud computing

Cloud computing offers a broad range of resource pools for conserving a huge quantity of information. Due to the intrusion of attackers, the information that exists in the cloud is threatened. Distributed Denial of Service (DDoS) attack is the main reason for attacks in the cloud. In this study, a Fractional Anti Corona Virus Optimization-based Deep Neuro-Fuzzy Network (FACVO-based DNFN) is devised for detecting DDoS in the cloud. The production of log files, feature fusion, data augmentation, and DDoS attack detection is the processing stages involved in this phase of the DDoS attack detection process. The feature fusion is carried out by RV coefficient and Deep Quantum Neural Network (Deep QNN), and the data augmentation is performed. Then, the Anti Corona Virus Optimization (ACVO) method and Fractional Calculus (FC) are both incorporated to create the FACVO algorithm. The DNFN is trained by the created FACVO algorithm, which identifies the DDoS attack. The proposed approach achieved testing accuracy, TPR, TNR, and precision values of 0.9304, 0.9088, 0.9293, and 0.8745 for using the NSL-KDD dataset without attack, and 0.9200, 0.8991, 0.9015, and 0.8648 for using the BoT-IoT dataset without attack.

Accelerated estimation of coffee sensory profiles using an AI-assisted electronic tongue

The use of a miniaturized potentiometric electronic tongue based on low-selective polymeric sensors was demonstrated for the discrimination and sensory characterization of coffee samples. The sensor array was able to discriminate 21 varieties of coffee with an average accuracy of 91.3% by combining hand-crafted features, predictor importance methods and trained classification models. Moreover, the e-tongue supported by a single regressor could be successfully trained to predict simultaneously the intensity of 13 coffee descriptors by leveraging dimensionality reduction to learn their interdependence. Sensory profiles of 33 samples were reconstructed with a 0.78 RV coefficient of agreement with sensory data using a rigorous leave-one-coffee-out validation. This study emphasizes the advantages of data-driven sensing approaches based on training by examples to help increase sample throughput when exploring new formulations and accelerate product design cycles.