Variety Of Metrics Research Articles

A multi-objective Immune Balancing Algorithm for Distributed Heterogeneous Batching-integrated Assembly Hybrid Flowshop Scheduling

Driven by economic globalization, global supply chain collaboration has gained significant importance, fostering the emergence of distributed manufacturing. This paper addresses the Distributed Heterogeneous Batching-integrated Assembly Hybrid Flowshop Scheduling (DHBIAHFS) problem within the pharmaceutical industry. Jobs are allocated to factories for processing, batched within defined lot sizes for transportation, and subsequently assembled into products to minimize the maximum completion time and tardy product count. Effective lot sizing during transport is emphasized between factories and assembly machines. Drawing inspiration from the biological immune system’s balancing mechanisms, we propose a Multi-objective Immune Balancing Algorithm (MOIBA) equipped with learning and repairing mechanisms. Each solution is structured with three nested sequences, and composite heuristic evaluations are employed to generate high-quality initial solutions. The performance of each solution is assessed based on both fitness and diversity metrics. Customized crossover and mutation operators are introduced with dynamically adjusted probabilities reflective of immune response dynamics. Quantitative analysis validates our mathematical model and the distinct components of MOIBA. We compare MOIBA’s efficiency against six other effective multi-objective strategies using three performance metrics. Stability and robustness assessments, conducted through variance examination and statistical testing, offer insights into MOIBA’s consistency and reliability across diverse problem instances.

Sacrificing sustainability for a higher GDP growth rate

Striving for positive GDP growth often comes at the expense of sustainability indicators, highlighting a crucial dilemma in modern economic development. This empirical study delves into the intricate relationship between sovereign sustainability indicators and GDP growth by using panel data that spans across 15 years for 100 countries and sheds light on the complexities of economic progress in the context of sustainability. Analysis reveals a negative bidirectional relationship between SDG scores and GDP growth rate, prompting further exploration into the determinants of these variables. Notably, GDP growth tends to rise following deteriorations in sustainability, while improvements in certain sustainability indicators, such as quality education and renewable energy consumption, correspond to enhanced GDP growth. However, the interplay between GDP growth factors and SDG scores unveils nuanced patterns, with both positive and negative associations observed across various determinants. These findings underscore the multifaceted nature of the relationship between sustainability and economic growth, challenging the notion that GDP growth alone signifies economic well-being. Considering these insights, policymakers are urged to adopt a holistic approach to sustainable development, integrating diverse econometric metrics, decoupling economic growth from environmental degradation, and bolstering data collection and analysis efforts. By embracing comprehensive strategies that prioritize sustainability alongside economic prosperity, policymakers can pave the way for inclusive and resilient development pathways.

Alterations of the microbiome across body sites in systemic lupus erythematosus: A systematic review and meta-analysis.

Systemic lupus erythematosus (SLE) is a complex autoimmune disease with unclear etiology. Growing evidence suggests the microbiome plays a role in SLE pathogenesis. However, findings are inconsistent across studies due to factors like small sample sizes and geographical variations. A comprehensive meta-analysis is needed to elucidate microbiome alterations in SLE. This study aimed to provide a systematic overview of microbiota dysbiosis across body sites in SLE through a meta-analysis of alpha diversity indices, beta diversity indices, and abundance taxa of microbiome. A literature search was conducted across four databases to identify relevant studies comparing SLE patients and healthy controls. Extracted data encompassed alpha and beta diversity metrics, as well as bacterial, fungal, and viral abundance across gut, oral, skin, and other microbiota. Study quality was assessed using the Newcastle-Ottawa Scale. Standardized mean differences and pooled effect sizes were calculated through meta-analytical methods. The analysis showed reduced alpha diversity and distinct beta diversity in SLE, particularly in the gut microbiota. Taxonomic analysis revealed compositional variations in bacteria from the gut and oral cavity. However, results for fungi, viruses, and bacteria from other sites were inconsistent due to limited studies. This meta-analysis offers a comprehensive perspective on microbiome dysbiosis in SLE patients across diverse body sites and taxa. The observed variations underscore the microbiome's potential role in SLE pathogenesis. Future research should address geographical variations, employ longitudinal designs, and integrate multi-omics approaches.

Comparison of the urinary microbiome in men who have sex with men with and without Chlamydia trachomatis infection

PurposeThe urinary tract is colonized by microbial communities that impact urinary health. Previous studies have suggested that the bacterial composition of the male urinary microbiota is related to STIs. This study assessed the bacterial composition of the urinary microbiome in South African MSM with and without C. trachomatis.MethodsThis study used urine samples from MSM attending care at the King Edward VIII hospital and the Aurum Institute in Durban, South Africa. A total of 200 samples were tested for C. trachomatis infection using the Applied Biosystems™ TaqMan® Assays. Urinary microbiomes of 23 samples were characterized using 16 S rRNA (V3 and V4) gene sequencing on the Illumina MiSeq platform.ResultsBacterial taxonomic analysis showed a high abundance of Streptococcus, Corynebacterium, and Staphylococcus in all the sequenced samples. Moreover, Prevotella and Lactobacillus were detected in urine samples of MSM. Alpha diversity metrics showed a slight increase in microbial diversity in C. trachomatis positive samples; however, this was not significant (ANOVA, P > 0.05). Principal coordinates analysis (PCoA) showed that the microbiome of C. trachomatis infected MSM was not clearly different from those uninfected. Distinct bacterial communities were not detected between positive and negative samples (PERMANOVA F1,22= 1.0284, R2 = 0.047%, P = 0.385).ConclusionMost microbiome studies on MSM to date have focused on the gut microenvironment. Few studies, however, have provided data regarding the normal composition of the male urethral microbiomes or if these microbiomes are associated with male STIs. This study adds to the growing body of knowledge highlighting the urinary microbiome in MSM.

Tree diversity drives understory carbon storage rather than overstory carbon storage across forest types

Although numerous studies have proposed explanations for the specific and relative effects of stand structure, plant diversity, and environmental conditions on carbon (C) storage in forest ecosystems, understanding how these factors collectively affect C storage in different community layers (trees, shrubs, and herbs) and forest types (mixed, broad-leaved (E), broad-leaved (M), and coniferous forest) continues to pose challenges. To address this, we used structural equation models to quantify the influence of biotic factors (mean DBH, mean height, maximum height, stem density, and basal area) and abiotic factors (elevation and canopy openness), as well as metrics of species diversity (Shannon–Wiener index, Simpson index, and Pielou’s evenness) in various forest types. Our analysis revealed the critical roles of forest types and elevation in explaining a substantial portion of variability in C storage in the overstory layer, with a moderate influence of stand factors (mean DBH and basal area) and a slightly negative impact of tree species diversity (Shannon–Wiener index). Notably, forest height emerged as the primary predictor of C storage in the herb layer. Regression relationships further highlighted the significant contribution of tree species diversity to mean height, understory C storage, and branch biomass within the forest ecosystem. Our insights into tree species diversity, derived from structural equation modeling of C storage in the overstory, suggest that the effects of tree species diversity may be influenced by stem biomass in statistical reasoning within temperate forests. Further research should also integrate tree species diversity with tree components biomass, forest mean height, understory C, and canopy openness to understand complex relationships and maintain healthy and sustainable ecosystems in the face of global climate challenges.

Capacity configuration of fuel cell hybrid vehicles using enhanced multi-objective particle swarm optimization with competitive mechanism

A well-designed hybrid powertrain is crucial for ensuring the safe, efficient, and durable operation of fuel cell hybrid vehicles. This paper introduces a modular design approach for powertrains, utilizing a Competitive mechanism-based Multi-Objective Particle Swarm Optimization (CMOPSO) algorithm integrated with nested dynamic programming. In this approach, the upper layer employs the CMOPSO algorithm to design the powertrain system, while the lower layer optimizes power coordination for each proposed design. This two-layer optimization framework considers factors such as vehicle economy and durability. Under WLTP conditions, the capacity configuration results are a fuel cell with a rated power of 22 kW, 100 batteries in series, and 7 batteries in parallel. Furthermore, the modular approach outperforms three other algorithms in terms of solution count, diversity, and overall performance metrics. The study also highlights that the vehicle’s power demand characteristics, influenced by different driving cycles, significantly affect capacity configuration results. Sensitivity analysis reveals that both the total operating cost and manufacturing cost of the vehicle are most sensitive to variations in the fuel cell rated power.

Predicting how varying moisture conditions impact the microbiome of dust collected from the International Space Station

BackgroundThe commercialization of space travel will soon lead to many more people living and working in unique built environments similar to the International Space Station, which is a specialized closed environment that contains its own indoor microbiome. Unintended microbial growth can occur in these environments as in buildings on Earth from elevated moisture, such as from a temporary ventilation system failure. This growth can drive negative health outcomes and degrade building materials. We need a predictive approach for modeling microbial growth in these critical indoor spaces.ResultsHere, we demonstrate that even short exposures to varying elevated relative humidity can facilitate rapid microbial growth and microbial community composition changes in dust from spacecraft. We modeled fungal growth in dust from the International Space Station using the time-of-wetness framework with activation and deactivation limited growth occurring at 85% and 100% relative humidity, respectively. Fungal concentrations ranged from an average of 4.4 × 106 spore equivalents per milligram of dust in original dust with no exposure to relative humidity to up to 2.1 × 1010 when exposed to 100% relative humidity for 2 weeks. As relative humidity and time-elevated increased, fungal diversity was significantly reduced for both alpha (Q < 0.05) and beta (R2 = 0.307, P = 0.001) diversity metrics. Bacteria were unable to be modeled using the time-of-wetness framework. However, bacterial communities did change based on constant relative humidity incubations for both beta (R2 = 0.22, P = 0.001) and alpha diversity decreasing with increasing moisture starting at 85% relative humidity (Q < 0.05).ConclusionOur results demonstrate that moisture conditions can be used to develop and predict changes in fungal growth and composition onboard human-occupied spacecraft. This predictive model can be expanded upon to include other spacecraft environmental factors such as microgravity, elevated carbon dioxide conditions, and radiation exposure. Understanding microbial growth in spacecraft can help better protect astronaut health, fortify spacecraft integrity, and promote planetary protection as human activity increases in low-Earth orbit, the moon, Mars, and beyond.5L5VMn4841emXVRzhjbC6HVideo

Transformer-Based Topic Modeling for Urdu Translations of the Holy Quran

Topic modeling enables the discovery of concealed themes and patterns in extensive text collections. It facilitates a thorough examination of the messages present in religious texts. Topic modeling for Quranic verses is a trending study area, with various translations already explored including Bahasa, English, and Arabic. Yet, there is a need for further research, particularly in Urdu translations of the Quran. In this study, we propose applying the BERTopic framework to Urdu translations of The Holy Quran. By leveraging the BERTopic approach, which incorporates a fine-tuned BERT model, we aim to capture the contextual nuances and linguistic complexities unique to the Quran. In this study, we utilized existing Urdu translations of the Quran from eight different translators sourced from Tanzil, a renowned resource for Quranic text and translations. We assessed the performance of our proposed BERTopic model compared to traditional techniques like LDA and NMF, using coherence and diversity metrics. The results indicate that our BERT-based approach outperforms these conventional methods, achieving an average coherence improvement of 0.03 and a diversity score of 0.83. These findings highlight the effectiveness of BERTopic in extracting meaningful topics from Urdu translations of The Holy Quran and contribute to the computational analysis of religious texts, supporting scholarly endeavours in comparative studies of Quranic translations in Urdu.

Previously unmeasured genetic diversity explains part of Lewontin's paradox in a -mer-based meta-analysis of 112 plant species.

At the molecular level, most evolution is expected to be neutral. A key prediction of this expectation is that the level of genetic diversity in a population should scale with population size. However, as was noted by Richard Lewontin in 1974 and reaffirmed by later studies, the slope of the population size-diversity relationship in nature is much weaker than expected under neutral theory. We hypothesize that one contributor to this paradox is that current methods relying on single nucleotide polymorphisms (SNPs) called from aligning short reads to a reference genome underestimate levels of genetic diversity in many species. To test this idea, we calculated nucleotide diversity ( ) and -mer-based metrics of genetic diversity across 112 plant species, amounting to over 205 terabases of DNA sequencing data from 27,488 individual plants. We then compared how these different metrics correlated with proxies of population size that account for both range size and population density variation across species. We found that our population size proxies scaled anywhere from about 3 to over 20 times faster with -mer diversity than nucleotide diversity after adjusting for evolutionary history, mating system, life cycle habit, cultivation status, and invasiveness. The relationship between -mer diversity and population size proxies also remains significant after correcting for genome size, whereas the analogous relationship for nucleotide diversity does not. These results suggest that variation not captured by common SNP-based analyses explains part of Lewontin's paradox in plants.

A holistic approach to software fault prediction with dynamic classification

Software Fault Prediction is a critical domain in machine learning aimed at pre-emptively identifying and mitigating software faults. This study addresses challenges related to imbalanced datasets and feature selection, significantly enhancing the effectiveness of fault prediction models. We mitigate class imbalance in the Unified Dataset using the Random-Over Sampling technique, resulting in superior accuracy for minority-class predictions. Additionally, we employ the innovative Ant-Colony Optimization algorithm (ACO) for feature selection, extracting pertinent features to amplify model performance. Recognizing the limitations of individual machine learning models, we introduce the Dynamic Classifier, a ground-breaking ensemble that combines predictions from multiple algorithms, elevating fault prediction precision. Model parameters are fine-tuned using the Grid-Search Method, achieving an accuracy of 94.129% and superior overall performance compared to random forest, decision tree and other standard machine learning algorithms. The core contribution of this study lies in the comparative analysis, pitting our Dynamic Classifier against Standard Algorithms using diverse performance metrics. The results unequivocally establish the Dynamic Classifier as a frontrunner, highlighting its prowess in fault prediction. In conclusion, this research introduces a comprehensive and innovative approach to software fault prediction. It pioneers the resolution of class imbalance, employs cutting-edge feature selection, and introduces dynamic ensemble classifiers. The proposed methodology, showcasing a significant advancement in performance over existing methods, illuminates the path toward developing more accurate and efficient fault prediction models.

Conceptual review of outcome metrics and measures used in clinical evaluation of artificial intelligence in radiology.

Artificial intelligence (AI) has numerous applications in radiology. Clinical research studies to evaluate the AI models are also diverse. Consequently, diverse outcome metrics and measures are employed in the clinical evaluation of AI, presenting a challenge for clinical radiologists. This review aims to provide conceptually intuitive explanations of the outcome metrics and measures that are most frequently used in clinical research, specifically tailored for clinicians. While we briefly discuss performance metrics for AI models in binary classification, detection, or segmentation tasks, our primary focus is on less frequently addressed topics in published literature. These include metrics and measures for evaluating multiclass classification; those for evaluating generative AI models, such as models used in image generation or modification and large language models; and outcome measures beyond performance metrics, including patient-centered outcome measures. Our explanations aim to guide clinicians in the appropriate use of these metrics and measures.

Sleeve Gastrectomy Protects Against Hypertension in Rats due to Changes in the Gut Microbiome

IntroductionSleeve gastrectomy (SG), results in improvement in hypertension. We have previously published that rodent SG improves hypertension independent of weight loss associated with unique shifts in the gut microbiome. We tested if the gut microbiome directly improves blood pressure by performing fecal material transfer (FMT) from post-SG rats to surgery-naïve animals. MethodsWe performed SG or Sham surgery in male, Zucker rats (n = 6-7) with obesity. Stool was collected postop from surgical donors for treatment of recipient rats. Three nonsurgical groups received daily, oral consumption of SG stool, sham stool, or vehicle alone (Nutella) for 10 wk (n = 7-8). FMT treatment was assessed for effects on body weight, food intake, oral glucose tolerance, and blood pressure. Genomic DNA of stool from donor and recipient groups were sequenced by 16S ribosomal RNA and analyzed for diversity, abundance, and importance. ResultsTen weeks of SG-FMT treatment significantly lowered systolic blood pressures in surgery-naïve, recipient rats compared to vehicle treatment alone (126.8 ± 13.3 mmHg versus 151.8 ± 12.2 mmHg, P = 0.001). SG-FMT treatment also significantly altered beta diversity metrics compared to Sham-FMT and vehicle treatment. In random forest analysis, amplicon sequence variant level significantly predicted FMT group, P = 0.01. ConclusionsWe have found a direct link between gut microbial changes after SG and regulation of blood pressure. Future mechanistic studies are required to learn what specific gut microbial changes are required to induce improvements in obesity-associated hypertension and translation to clinical, metabolic surgery.

Impact of the 2011 Tohoku Earthquake on the species diversity of rocky intertidal sessile assemblages.

The impacts of large-scale disturbance events on the species diversity of rocky intertidal sessile assemblages across multiple spatial scales are not well understood. To evaluate the influence of the 2011 Tohoku Earthquake on alpha and beta diversities of rocky intertidal sessile assemblages, we surveyed sessile assemblages in the mid-shore zone from 2011 to 2019 and compared the data with those collected from 2003 to 2010 before the earthquake at the same region. The census was conducted across 22 study plots on five rocky shores along 30 km of the Sanriku Coast of Japan, which is located 150-160 km north-northwest of the earthquake epicenter. Alpha diversity was measured with three Hill numbers (H 0, H 1, and H 2), which represent the number of equally common species that would exist in a community with the same diversity as the sampled community, with higher values of the subscript indicating more weight placed on abundant species. Beta diversity was measured with two metrics (BD total at two spatial scales). Values were compared between the post-earthquake period (2011-2019) and the pre-earthquake period (2003-2010). The results show that the Tohoku Earthquake significantly altered the species diversity of intertidal sessile assemblages across multiple spatial scales. All diversity metrics obtained at multiple spatial scales (i.e., alpha diversities: H 0, H 1, and H 2; beta diversities: BD total at the shore and regional scales) decreased immediately after the earthquake and then increased in subsequent years. At 2 years after the earthquake, H 0 recovered to within the range of pre-earthquake values and H 1 and H 2 became significantly higher than pre-earthquake values. Most metrics of alpha and beta diversities recovered to pre-earthquake levels after several years, but regional BD total remained low for a longer period.

Beta diversity subcomponents of plant species turnover and nestedness reveal drivers of community assembly in a regenerating subtropical forest.

Secondary forests represent a significant proportion of global forest cover, with over 70% of forests in East Asia classified as regenerating. While succession has been studied extensively in temperate systems, trajectories of subtropical succession remain poorly characterized in highly disturbed, urban-adjacent forests. Investigating the additive beta diversity components of turnover and nestedness may reveal community assembly mechanisms driving secondary succession. The present study investigates plant community assembly along a successional gradient from 7 to 70 years following the onset of succession in secondary subtropical forests in Hong Kong, China. Plant survey data for 28 plots were analysed, generating additive Simpsons turnover and nestedness beta diversity metrics. Dissimilarity matrices were generated and modelled as a function of environmental matrices including forest plant community age (years following onset of secondary succession), inter-community distance (metres), and soil moisture saturation (%) across three elevational bands using generalized dissimilarity models. Nonmetric multidimensional scaling of plant communities was conducted with Bray-Curtis dissimilarity matrices. Inter-community distance and successional age differentially influenced plant species turnover between lowland and Montane forest types. Models of nestedness found that plot age and soil moisture saturation were significant drivers of nestedness patterns in plant communities across elevational classes. Turnover represented a higher proportion of Sorensen beta diversity than nestedness, while ANOSIM found significant differentiation between plant communities at different successional stages. Turnover patterns suggest a deterministic model of community assembly, with strong patterns of species replacement between communities at fine spatial scales and successional stages, as well as clear compositional shifts between lowland and montane forest types. NMDS analysis and functional compositional assessments suggested a transition from early successional communities with a high proportion of shrub species, to later successional communities with a higher proportion of tree species, with an increase in species turnover with greater age dissimilarity.

The classified peri-urban forest of Kua from Burkina Faso must not be declassified for ecological reasons

Densification of urban areas and rapid urbanization result in land scarcity in the Sahelian cities of West Africa. Therefore, urban greenspaces are increasingly coveted for socio-economic development. This is the case of the peri-urban forest of Kua in Burkina Faso which has been subject of controversies between ecologists and decision makers who intended to declassify this forest for hospital project. The lack of ecological evidence on this forest fueled the dichotomous conflicts between environment protection and urban planning. Hence, this study aimed to assess the ecological characteristics of the peri-urban forest of Kua. Dendrometric measurements of 2993 individuals were performed in 106 plots across the peri-urban forests of Kua and Dindéresso. We computed and compared tree diversity metrics, structural parameters and carbon stocks between the two forests. The findings revealed a γ–diversity of 40 woody species from 38 genera and 16 families in Kua forest. Through the diversity metrics, we found that the peri-urban forest of Kua exhibited lower tree diversity than the peri-urban forest of Dindéresso (p-value < 0.05). Similarly, tree density was significantly lower (p-value < 0.05) in Kua (212.446 ± 291.65 ind. ha−1) than in Dindéresso (378.461 ± 298.89 ind. ha−1). However, carbon stock was two times significantly higher (p-value < 0.05) in Kua (26.21 ± 31.27 Mg.ha−1) than in Dindéresso (13.389 ± 15.95 Mg.ha−1), highlighting the high potential of this forest in carbon sequestration. This study demonstrated that the peri-urban forest of Kua – despite its lower tree diversity and density – constitutes important carbon pools and sinks. Therefore, this forest must not be declassified, but should rather be reinforced through management plans and conservation actions. To sustain urban development, land management policies promoting the conservation of greenspaces are required to reduce biodiversity loss and carbon emissions in urban centers.

In search of per capita effects of Prunus serotina Ehrh. invasion on temperate forest understory alpha diversity

Prunus serotina is one of the most widespread alien tree species in temperate European forests. Although numerous studies revealed both negative and positive impacts on native ecosystems, only a few assessed the quantitative impacts along a species abundance gradient, based on aboveground biomass of per capita effects. Here we studied how alpha diversity of forest understory changes with increasing aboveground biomass of P. serotina. In Wielkopolska National Park (W Poland) we established 92 plots in both nutrient-rich and poor Pinus sylvestris stands. We assessed the effects of P. serotina proportion on understory species composition using non-metric multidimensional scaling (NMDS). We also assessed the effects of P. serotina proportion on alpha diversity metrics using generalized mixed-effects linear models, accounting also for stand age, soil C:N ratio, and light availability. P. serotina biomass ranged from 0.0 to 34.29 Mg ha−1 (with an average of 5.70 ± 0.83 Mg ha−1), comprising 0% to 17.26% (with an average of 2.93 ± 0.43%) of total woody species biomass. NMDS revealed a lack of P. serotina proportion effects on understory species composition. We found small positive impacts of P. serotina proportion on total species richness (+ 8.8 species along a gradient with a 10% increase of P. serotina biomass proportion) and functional richness (+ 0.08). The positive impacts of P. serotina were mostly related to the confounding effect of two factors: an increase in light availability facilitates both the development of understory plants as well as the growth of P. serotina.

Mitochondrial DNA control-region and coding-region data highlight geographically structured diversity and post-domestication population dynamics in worldwide donkeys.

Donkeys (Equus asinus) have been used extensively in agriculture and transportations since their domestication, ca. 5000-7000 years ago, but the increased mechanization of the last century has largely spoiled their role as burden animals, particularly in developed countries. Consequently, donkey breeds and population sizes have been declining for decades, and the diversity contributed by autochthonous gene pools has been eroded. Here, we examined coding-region data extracted from 164 complete mitogenomes and 1392 donkey mitochondrial DNA (mtDNA) control-region sequences to (i) assess worldwide diversity, (ii) evaluate geographical patterns of variation, and (iii) provide a new nomenclature of mtDNA haplogroups. The topology of the Maximum Parsimony tree confirmed the two previously identified major clades, i.e. Clades 1 and 2, but also highlighted the occurrence of a deep-diverging lineage within Clade 2 that left a marginal trace in modern donkeys. Thanks to the identification of stable and highly diagnostic coding-region mutational motifs, the two lineages were renamed as haplogroup A and haplogroup B, respectively, to harmonize clade nomenclature with the standard currently adopted for other livestock species. Control-region diversity and population expansion metrics varied considerably between geographical areas but confirmed North-eastern Africa as the likely domestication center. The patterns of geographical distribution of variation analyzed through phylogenetic networks and AMOVA confirmed the co-occurrence of both haplogroups in all sampled populations, while differences at the regional level point to the joint effects of demography, past human migrations and trade following the spread of donkeys out of the domestication center. Despite the strong decline that donkey populations have undergone for decades in many areas of the world, the sizeable mtDNA variability we scored, and the possible identification of a new early radiating lineage further stress the need for an extensive and large-scale characterization of donkey nuclear genome diversity to identify hotspots of variation and aid the conservation of local breeds worldwide.

Plant phylogenetic diversity along the urban–rural gradient and its association with urbanization degree in Shanghai, China

ContextThe spatial distribution of plant diversity in urban areas is fundamental to understanding the relationship between urbanization and biodiversity. Previous research has primarily focused on taxonomic levels to assess species richness. In contrast, investigations into the spatial patterns of phylogenetic diversity in urban plants remain limited.ObjectivesThis study aims to investigate the spatial patterns of plant phylogenetic diversity along an urban–rural gradient and quantify how phylogenetic diversity and the degree of urbanization are related.MethodsA survey of vascular plants was conducted at 134 randomly selected sample plots along four urban–rural transects in Shanghai, China. Three phylogenetic diversity metrics were calculated: Faith’s phylogenetic diversity (PD), net relatedness index (NRI), and net nearest taxon index (NTI), along with the urbanization degree index (UDI). Regression analysis was employed to quantify the spatial patterns of plant phylogenetic diversity across different taxa along the urban–rural gradients and their relationships with UDI.ResultsThe study observed seven distinct patterns of plant phylogenetic diversity along the urban–rural gradients in different taxa, which support the previous hypotheses that biological distribution patterns at the species level also hold true at the phylogenetic level. Faith’s phylogenetic diversity (PD) showed a linear increase with increasing UDI for total, woody, perennial, and cultivated plant assemblages. The UDI explained 3–36% of the variation in PD for these taxa. In contrast, PD for annual and spontaneous plants exhibited a linear decrease with increasing UDI, which explained 25% and 3% variation in PD for annual and spontaneous plants, respectively. The net relatedness index (NRI) for woody, perennial, and cultivated plants, as well as the net nearest taxon index (NTI) for perennial and cultivated plants, linearly increases with UDI, whereas the NRI for total, annual, and spontaneous plants, as well as NTI for total, woody, annual, and spontaneous plants linearly decrease with UDI. However, some of these trends were only marginally significant.ConclusionsThe spatial patterns of plant phylogenetic diversity varied along the urban-to-rural gradients, indicating that urban environmental filtering has an impact on plant phylogenetic diversity. Urbanization increased the phylogenetic richness of different plant taxa in Shanghai but resulted in more clustering and relatedness of species within plant assemblages. Phylogenetic richness exhibited a linear increase with UDI, while the phylogenetic divergence decreased with UDI. The UDI is a useful predictor for examining variations in plant phylogeny due to urbanization. Our findings provide insights into how urbanization impacts plant phylogenetic diversity, helping urban plant diversity conservation.

IMUGPT 2.0: Language-Based Cross Modality Transfer for Sensor-Based Human Activity Recognition

One of the primary challenges in the field of human activity recognition (HAR) is the lack of large labeled datasets. This hinders the development of robust and generalizable models. Recently, cross modality transfer approaches have been explored that can alleviate the problem of data scarcity. These approaches convert existing datasets from a source modality, such as video, to a target modality, such as inertial measurement units (IMUs). With the emergence of generative AI models such as large language models (LLMs) and text-driven motion synthesis models, language has become a promising source data modality as well - as shown in proof of concepts such as IMUGPT. In this work, we conduct a large-scale evaluation of language-based cross modality transfer to determine their effectiveness for HAR. Based on this study, we introduce two new extensions for IMUGPT that enhance its use for practical HAR application scenarios: a motion filter capable of filtering out irrelevant motion sequences to ensure the relevance of the generated virtual IMU data, and a set of metrics that measure the diversity of the generated data facilitating the determination of when to stop generating virtual IMU data for both effective and efficient processing. We demonstrate that our diversity metrics can reduce the effort needed for the generation of virtual IMU data by at least 50%, which opens up IMUGPT for practical use cases beyond a mere proof of concept.

Diffusion-based diverse audio captioning with retrieval-guided Langevin dynamics

Audio captioning, a comprehensive task of audio understanding, aims to provide a natural-language description of an audio clip. Beyond accuracy, diversity is also a critical requirement for this task. Human-produced captions possess rich variability due to the ambiguity of audio semantics (such as insects buzzing and electrical humming making similar sounds) and the existence of subjective judgments (metaphor, affections, etc.). However, current diverse audio captioning systems fail to produce captions with near-human diversity. Recently, diffusion models have demonstrated the potential to generate data with diversity while maintaining decent accuracy, yet they have not been explored in audio captioning. On the other hand, diffusion models tend to have low generation accuracy and fluency on text data. Directly applying diffusion models to audio captioning tasks may aggravate this problem due to the small size of annotated datasets and the mutable supervision target caused by the variability in human-produced captions. In this work, we propose a diffusion-based diverse audio captioning model by incorporating the BART language model into the diffusion model for better utilization of the pre-trained linguistic knowledge. We also propose a retrieval-guided Langevin dynamics module, which enables dynamic run-time alignment between generated captions and the target audio. Extensive experiments on standard audio captioning benchmark datasets (Clotho and AudioCaps) demonstrate that our model can achieve better performance on both accuracy and diversity metrics compared with state-of-the-art diverse audio captioning systems. The implementation is available at https://github.com/AI4MyBUPT/DACRLD.