Search Results (101)

Dog bite incidents constitute significant public health, animal welfare, and economic concerns with substantial physical and psychological consequences for victims. Despite legislative responses, research indicates that breed-focused interventions are ineffective in reducing dog bite risk. Human behaviour, caregiving practices, and environmental context all play central roles in the expression of human-directed canine aggression. Current methods of assessing dog bite risk remain largely unstructured, dog-centred, and reliant on subjective judgement and provocative behavioural testing. These approaches exhibit limited predictive validity and poor reliability, and are vulnerable to bias, raising serious concerns for public safety, judicial fairness, and animal welfare. Comparable challenges in human violence risk assessment led to the development of an evidence-based structured professional judgement (SPJ) assessment framework, which combines empirical risk factors with individualised case formulation and dynamic risk management. An SPJ framework for dog bite risk would ensure the systematic consideration of empirically supported static and dynamic risk factors relating to the dog, caregivers, and related environmental conditions, while supporting the development of targeted risk reduction strategies. Conclusion: Developing an SPJ approach offers a more scientifically grounded, ethically defensible, and prevention-focused method for managing dog bite risk, with potential benefits for public safety, animal welfare, and professional practice. Full article

Agricultural credit rationing remains a persistent systemic friction driven by information opacity and limited collateral. This study develops a credit risk early-warning system by fusing multi-source institutional digital footprints (tax compliance signals, judicial enforcement records, and credit history indicators) for 1021 agricultural enterprises in China. Methodologically, we propose a Default Event Isolation protocol to enforce strict ex ante validity by discarding observations at and after the event month, and implement a two-step feature optimization pipeline that reduces 138 predictors to a parsimonious set of 50 features. Empirically, the optimized LightGBM (version 4.6.0) model achieves an AUC = 0.9345 (95% bootstrap CI: 0.8745–0.9563) and PR-AUC = 0.4421, representing a 47× lift over the random baseline under extreme class imbalance (0.94% event rate), and captures 87.4% of early-warning events by monitoring only the top 10% highest-risk firms. The interpretability analysis consistently highlights judicial boundary constraints and tax stability signals as dominant predictors, forming a “judicial baseline + tax stability” dual-core structure. A strict credit-only robustness check using bank-recorded NPL labels maintains strong predictive performance (AUC = 0.9089, 95% bootstrap CI: 0.8255–0.9591), mitigating concerns that the model’s signal is driven by label overlap. These findings suggest that integrating institutional records into automated screening pipelines can enable the earlier and more targeted identification of distressed borrowers in rural lending, even when traditional financial statements are unavailable. Full article

The integration of artificial intelligence into criminal sentencing decisions represents one of the most consequential applications of algorithmic systems in contemporary governance. While AI-assisted risk assessment tools promise enhanced consistency and predictive accuracy, their deployment in judicial contexts raises profound concerns regarding transparency, due process, and fundamental rights. This paper proposes a comprehensive framework for AI-assisted sentencing modeling that embeds explainability as a foundational constraint rather than an afterthought. Drawing upon the landmark State v. Loomis decision, empirical analyses of the COMPAS algorithm, and emerging regulatory frameworks including the European Union Artificial Intelligence Act, we examine the tension between predictive performance and interpretive transparency. Our framework integrates a three-layer explanation architecture: inherent interpretability through generalized additive models (GA²Ms) providing transparent global structure, exact local feature attribution derived directly from the additive model decomposition without approximation, and counterfactual reasoning that identifies minimal input changes altering risk classifications. We demonstrate through rigorous experimental validation on the ProPublica COMPAS dataset (n = 6172) that explainability-constrained models achieve comparable predictive validity to opaque alternatives (AUC 0.71 versus 0.70–0.72 for black-box methods) while satisfying constitutional due process requirements and emerging regulatory mandates under the EU Artificial Intelligence Act. The impossibility theorems governing algorithmic fairness are examined in light of their implications for sentencing equity, and we propose that transparent model architectures enable targeted interventions unavailable when decision logic remains concealed. The paper concludes with policy guidance for jurisdictions seeking to implement AI-assisted sentencing systems that balance public safety objectives with procedural fairness and individual rights. Full article

Background: Dietary guidance for type 2 diabetes mellitus (T2DM) frequently discourages fruit consumption due to intrinsic sugars, despite extensive evidence supporting the anti-diabetic properties of fruit-derived polyphenols. This reductionist, carbohydrate-only model inadequately reflects the complex bioactive matrices of whole fruits. Objective: To develop an integrated analytical framework that quantitatively balances the predicted anti-diabetic bioactivity of fruit polyphenols against their glycemic burden, and to apply this model to fruits commonly consumed in the Korean market. Methods: Nutritional and phytochemical composition data for five fruits sourced from Korea and India were obtained from national food databases to ensure broader phytochemical diversity. Six representative polyphenols were selected based on abundance and reported significance. Molecular docking was conducted against α-glucosidase and peroxisome proliferator-activated receptor gamma (PPAR-γ) to estimate target-specific affinity, and a “Total Predicted Anti-Diabetic Score” (TPAS) was computed by weighting docking potency by compound concentration. A risk–benefit matrix integrating TPAS and sugar content was applied to classify fruits, followed by a cultivar-level comparison of major grape varieties. Results: Hesperidin exhibited the strongest predicted PPAR-γ binding (−9.3 kcal/mol). Among whole fruits, grapes and oranges showed the highest TPAS (593.19 and 448.77, respectively), placing them in the “redemptive choice” category (high benefit/high glycemic risk). Comparative cultivar analysis identified the Campbell Early grape as the most advantageous option, with a Holistic Value Score (HVS) of 9.5, notably higher than Shine Muscat (3.9). Conclusions: This study presents a structured, computation-driven framework capable of integrating phytochemical potency and nutritional risk into a unified metric for dietary evaluation. Despite higher sugar content, fruits rich in potent polyphenols may confer substantial metabolic benefit when consumed judiciously. These findings challenge sugar-centric dietary models and provide an evidence-based tool for consumer-level guidance in T2DM dietary management. Full article

Modeling legal reasoning with artificial intelligence and machine learning presents formidable challenges. Legal decisions emerge from a complex interplay of factual circumstances, statutory interpretation, case precedent, jurisdictional variation, and human judgment—including the behavioral characteristics of judges and juries. This paper takes an exploratory approach to investigating how contemporary ML techniques might capture aspects of this complexity. Using pharmaceutical patent litigation as an illustrative domain, we develop a multi-layer analytical pipeline integrating text mining, clustering, topic modeling, and classification to analyze 698 U.S. federal district court decisions spanning January 2016 through December 2018, comprising substantive validity and infringement rulings under the Hatch-Waxman regulatory framework. Results demonstrate that the pipeline achieves 85–89% prediction accuracy—substantially exceeding the 42% baseline majority-class rate and comparing favorably with prior legal prediction studies—while producing interpretable intermediate outputs: clusters that correspond to recognized doctrinal categories (Abbreviated New Drug Application—ANDA litigation, obviousness, written description, claim construction) and topics that capture recurring legal themes. We discuss what these findings reveal about both the possibilities and limitations of computational approaches to legal reasoning, acknowledging the significant gap between statistical prediction and genuine legal understanding. Full article

Legal judgment prediction (LJP) increasingly relies on large language models whose full fine-tuning is memory-intensive and susceptible to catastrophic forgetting. We present LawLLM-DS, a two-stage Low-Rank Adaptation (LoRA) framework that first performs legal knowledge pre-tuning with an aggressive learning rate and subsequently refines judgment relations with conservative updates, using dedicated LoRA adapters, 4-bit quantization, and targeted modification of seven Transformer projection matrices to keep only 0.21% of parameters trainable. From a structural perspective, the twenty annotated legal elements form a symmetric label co-occurrence graph that exhibits both cluster-level regularities and asymmetric sparsity patterns, and LawLLM-DS implicitly captures these graph-informed dependencies while remaining compatible with downstream GNN-based representations. Experiments on 5096 manually annotated divorce cases show that LawLLM-DS lifts macro F1 to 0.8893 and achieves an accuracy of 0.8786, outperforming single-stage LoRA and BERT baselines under the same data regime. Ablation studies further verify the contributions of stage-wise learning rates, adapter placement, and low-rank settings. These findings demonstrate that curriculum-style, parameter-efficient adaptation provides a practical path toward lightweight yet structure-aware LJP systems for judicial decision support. Full article

Artificial intelligence (AI) is transforming numerous sectors, and its integration into the legal domain holds significant potential for automating labor-intensive tasks, enhancing judicial decision-making, and improving overall system efficiency. This study introduces an AI-powered model, named the Legal Judgment Prediction Ensemble (LJPE), which is designed to predict legal case outcomes by leveraging historical judicial data. By using natural language processing (NLP) techniques, feature engineering, and a complex two-level stacking ensemble, the LJPE model has better predictive accuracy at 94.68% compared to modern legal language and conventional machine learning models. Moreover, the findings underline the predictive strength of textual features obtained from case facts, vote margins, and legal-specific features. This study offers a solid technical solution for predicting legal judgments for the responsible use of the model, helping to create a more efficient, transparent, and fair legal system. Full article

Invasive candidiasis (IC) is a life-threatening fungal infection predominantly affecting critically ill patients in intensive care units (ICUs). Despite advances in antifungal therapies, IC remains a diagnostic and therapeutic challenge, with a mortality rate exceeding 40%. The current reliance on blood cultures as the diagnostic gold standard is limited by low sensitivity and prolonged turnaround times, often delaying effective treatment. This often leads to the overuse of empirical antifungal therapies, increasing resistance, healthcare costs, and inconsistent outcomes. To address these issues, this paper introduces a five-step diagnostic strategy developed by an expert panel to optimise IC diagnosis and management. The strategy integrates predictive risk scores, biomarkers, and antifungal susceptibility testing to streamline diagnosis, identify high-risk patients, and promote antifungal stewardship. It also addresses barriers such as resource disparities and variability in clinical practices, offering a practical, standardised strategy for ICUs in the UK and Ireland. The clinical utility of this approach is highlighted through two patient cases. One describes the safe discontinuation of antifungal therapy after a negative (1,3)-β-D-glucan (BDG) assay ruled out IC, reducing unnecessary treatment and adverse effects. The other showcases the use of rapid in-house antifungal susceptibility testing to precisely tailor therapy for a patient with Nakaseomyces glabratus, ensuring effective treatment and preventing resistance. This pragmatic five-step guide simplifies and standardises IC diagnosis, aiming to lower mortality, optimise therapies, and promote judicious antifungal use. Full article

Colibacillosis, caused by avian pathogenic Escherichia coli (APEC), represents a major threat to poultry production, leading to significant mortality and economic losses. This study aimed to characterize an APEC strain, HPVN24, isolated from diarrheic chickens at a farm in Hai Phong, Vietnam. The strain was investigated through phenotypic assays, antibiotic susceptibility profiling, and whole-genome sequencing using the Illumina platform. HPVN24 exhibited β-hemolytic activity and resistance to trimethoprim, ampicillin, and ciprofloxacin. Whole-genome analysis identified the strain as serotype O78:H9 and sequence type ST23, with a genome size of 5.05 Mb and a GC content of 50.57%. Genome annotation revealed a wide repertoire of genes involved in metabolism, secretion systems, virulence, and biofilm formation. Virulence-associated genes included those related to adhesion, iron acquisition, hemolysin production, and stress response. Analysis predicted multidrug resistance to 18 antibiotic classes, with particularly strong resistance to fluoroquinolones. Phylogenetic comparison demonstrated that HPVN24 clustered closely with O78:H9 strains isolated from poultry in other regions, suggesting potential transmission across populations. These findings indicate that HPVN24 is a multidrug-resistant and highly virulent APEC strain linked to colibacillosis outbreaks in Vietnam and highlight the need for ongoing surveillance, judicious antibiotic usage, and alternative strategies to ensure poultry health and food safety. Full article

Legal judgment prediction is an emerging application of artificial intelligence in the legal domain, offering significant potential to enhance legal decision support systems. Such systems can improve judicial efficiency, reduce burdens on legal professionals, and assist in early-stage case assessment. This study focused on predicting whether a legal case would be Accepted or Rejected using only the Fact section of court rulings. A key challenge lay in processing long legal documents, which often exceeded the input length limitations of transformer-based models. To address this, we proposed a two-step methodology: first, each document was segmented into sentence-level inputs compatible with AraBERT—a pretrained Arabic transformer model—to generate sentence-level predictions; second, these predictions were aggregated to produce a document-level decision using several methods, including Mean, Max, Confidence-Weighted, and Positional aggregation. We evaluated the approach on a dataset of 19,822 real-world cases collected from the Saudi Arabian Commercial Court. Among all aggregation methods, the Confidence-Weighted method applied to the AraBERT-based classifier achieved the highest performance, with an overall accuracy of 85.62%. The results demonstrated that combining sentence-level modeling with effective aggregation methods provides a scalable and accurate solution for Arabic legal judgment prediction, enabling full-length document processing without truncation. Full article

Objective: U.S. courts have implemented interventions educating jurors about implicit bias, although evidence for their effectiveness remains limited. We explored public perceptions of these interventions that might influence their ability to improve trial fairness and identified psychological factors predicting such perceptions. Hypotheses: We hypothesized that certain psychological factors (i.e., political conservatism, psychological reactance, skepticism toward social scientists, implicit and explicit racial bias, advantaged-group identity management strategies) would predict support for implicit bias interventions in courts. Method: White participants (N = 1016)—some of whom watched an implicit bias intervention in one of two formats (educational video, judicial instructions)—provided their perceptions of implicit bias interventions, evaluated the intervention they watched (if applicable), and completed individual difference measures. Results: Overall, participants supported implicit bias interventions in both formats. However, political conservatism and other hypothesized individual difference measures were associated with less favorable perceptions. We further explored participants’ perspectives via a thematic content analysis of open-ended impressions of the interventions. Conclusions: Courts are adopting implicit bias interventions despite mixed research regarding their effectiveness and a limited understanding of how they are perceived. Our findings suggest that White participants generally favor these interventions and offer insight into the nuances of their perceptions. Full article

With the rapid growth of job-related crimes in developing economies, there is an urgent need for intelligent judicial systems to standardize sentencing practices. This study proposes a Multi-Level Feature Fusion Network (MLFFN) to enhance the accuracy and interpretability of sentencing predictions in job-related crime cases. The model integrates hierarchical legal feature representation, beginning with benchmark judgments (including starting-point penalties and additional penalties) as the foundational input. The frontend of MLFFN employs an attention mechanism to dynamically fuse word-level, segment-level, and position-level embeddings, generating a global feature encoding that captures critical legal relationships. The backend utilizes sliding-window convolutional kernels to extract localized features from the global feature map, preserving nuanced contextual factors that influence sentencing ranges. Trained on a dataset of job-related crime cases, MLFFN achieves a 6%+ performance improvement over the baseline models (BERT-base-Chinese, TextCNN, and ERNIE) in sentencing prediction accuracy. Our findings demonstrate that explicit modeling of legal hierarchies and contextual constraints significantly improves judicial AI systems. Full article

This paper presents a theoretical model of a thin, penny-shaped piezoelectric actuator bonded to an isotropic thermo-elastic substrate under coupled electrical-thermal-diffusive loading. The problem is assumed to be axisymmetric, and the peeling stress of the film is neglected in accordance with membrane theory, yielding a simplified equilibrium equation for the piezoelectric film. By employing potential theory and the Hankel transform technique, the surface strain of the substrate is analytically derived. Under the assumption of perfect bonding, a governing integral equation is established in terms of interfacial shear stress. The solution to this integral equation is obtained numerically using orthotropic Chebyshev polynomials. The derived results include the interfacial shear stress, stress intensity factors, as well as the radial and hoop stresses within the system. Finite element analysis is conducted to validate the theoretical predictions. Furthermore, parametric studies elucidate the influence of material mismatch and actuator geometry on the mechanical response. The findings demonstrate that, the performance of the piezoelectric actuator can be optimized through judicious control of the applied electrical-thermal-diffusive loads and careful selection of material and geometric parameters. This work provides valuable insights for the design and optimization of piezoelectric actuator structures in practical engineering applications. Full article

The justice system has increasingly applied AI techniques for legal judgment to enhance efficiency. However, most AI techniques focus on decision-making outcomes, failing to capture the deliberative nature of the real-world judicial process. To address these challenges, we propose a large language model-based multi-agent framework named AgentsBench. Our approach leverages multiple LLM-driven agents that simulate the discussion process of the Chinese judicial bench, which is often composed of professional and lay judge agents. We conducted experiments on a legal judgment prediction task, and the results show that our framework outperforms existing LLM-based methods in terms of performance and decision quality. By incorporating these elements, our framework reflects real-world judicial processes more closely, enhancing accuracy, fairness, and societal consideration. While the simulation is based on China’s lay judge system, our framework is generalizable and can be adapted to various legal scenarios and other legal systems involving collective decision-making processes. Full article

Legal actors often discuss emotion-based decisions and reasoned evaluation of the facts as distinct and opposite methods through which jurors can reach conclusions. However, research suggests that emotion can have an indirect effect on juror decisions by changing the way that jurors evaluate the facts of the case. In three studies (N = 713, N = 677, N = 651), we tested whether mock jurors’ negative moral emotions towards the defendant predicted their evaluations of unrelated case evidence and in turn their case judgments and whether judicial rehabilitation could reduce this effect. Participants read a civil case and were randomly assigned to either receive judicial rehabilitation or not. Then, they completed measures relating to their negative moral emotions towards the defendant, their agreement with plaintiff and defense evidence, and case judgments. When participants reported increased negative emotions towards the defendant, they agreed more with unrelated plaintiff evidence and less with unrelated defense evidence. In turn, they voted liable more often and awarded more in damages. Judicial rehabilitation did not reduce this effect. This research provides support for the idea that there is a more complicated relationship between emotion and decisions than legal actors suggest. Specifically, negative emotions towards the defendant are associated with a pro-plaintiff evaluation of evidence and pro-plaintiff judgments. Full article