Search Results (909)

Cyberbullying and hate-driven behavior on social media have become increasingly prevalent, posing serious psychological and social risks. This study proposes a machine learning-based approach to detect hate-driven content by integrating temporal and behavioral features—such as message frequency, interaction duration, and user activity patterns—alongside traditional text-based features. Furthermore, we extend our evaluation to include recent neural network architectures, namely ALBERT and BiLSTM, enabling a more robust representation of semantic and sequential patterns. Building on our previous research presented at JNIC-2024, we conduct a comparative evaluation of multiple classification algorithms using both existing and engineered datasets. The results show that incorporating non-textual features significantly improves detection accuracy and robustness. This work contributes to the development of intelligent cyberbullying detection systems and highlights the importance of behavioral context in online threat analysis. Full article

Ports serve as key nodes in eco-friendly and digital logistics networks, and the volume of cargo handled continues to increase in response to growing international trade. However, the increased workload within limited spaces heightens the risk of safety accidents, and the number of casualties in port stevedoring operations has continued to rise. As the era of transition toward eco-friendly and digital paradigms unfolds, the adoption of new technologies in ports presents a strategic opportunity to enhance safety management. As of 13 May 2025, the study conducted a text-mining analysis based on research abstracts related to the keyword “New technology and port safety,” in the context of internal and external environmental changes. Specifically, a total of 639 research abstracts were collected, but 138 abstracts, which were unrelated to port safety, were excluded, and 501 abstracts from the Clarivate Web of Science database were analyzed, focusing on 2676 words that appeared at least twice. The study applied Term Frequency (TF) analysis, TF–Inverse Document Frequency analysis, Semantic Network Analysis, and Topic Modeling. The results indicate that Internet of Things emerged as a core solution for strengthening port safety management. However, challenges remain, including the prevention of security breaches, high infrastructure implementation costs, and limitations in battery life. Full article

Tables, as a form of structured or semi-structured data, are widely found in documents, reports, and data manuals. Table-based question answering (TableQA) plays a key role in table document analysis and understanding. Existing approaches to TableQA can be broadly categorized into content-matching methods and end-to-end generation methods based on encoder–decoder deep neural networks. Content-matching methods return one or more table cells as answers, thereby preserving the original data and making them more suitable for downstream tasks. End-to-end methods, especially those leveraging large language models (LLMs), have achieved strong performance on various benchmarks. However, the variability in LLM-generated expressions and their heavy reliance on prompt engineering limit their applicability where answer fidelity to the source table is critical. In this work, we propose CBCM (Cell-by-Cell semantic Matching), a fine-grained cell-level matching method that extends the traditional row- and column-matching paradigm to improve accuracy and applicability in TableQA. Furthermore, based on the public IM-TQA dataset, we construct a new benchmark, IM-TQA-X, specifically designed for the multi-row and multi-column cell recall task, a scenario underexplored in existing state-of-the-art content-matching methods. Experimental results show that CBCM improves overall accuracy by 2.5% over the latest row- and column-matching method RGCNRCI (Relational Graph Convolutional Networks based Row and Column Intersection), and boosts accuracy in the multi-row and multi-column recall task from 4.3% to 34%. Full article

The rise of digital platforms has transformed heritage interpretation from a single official narrative to multi-stakeholder participation. This study investigates how such platforms mediate the formation of a sense of place at the Kulangsu World Heritage Site (WHS). Data were collected from official narrative texts and user-generated content (UGC) on Dianping and Ctrip, and analyzed using high-frequency word statistics and semantic network analysis. The results reveal a clear divergence between official narratives, which emphasize Outstanding Universal Value (OUV), and tourist perceptions, which focus on visual landmarks and “check-in” practices shaped by the “digital gaze.” Moreover, the sense of place is shown to be a dynamic process, co-constructed through pre-visit expectations, on-site experiences, and post-visit reflections. The findings also highlight a transformation in tourists’ roles, shifting from passive cultural consumers to active participants in the co-construction of heritage values, with digital platforms serving as critical mediators. Theoretically, the study advances digital heritage scholarship by clarifying the mechanism of the digital gaze and the dynamic nature of sense of place. Practically, it underscores the importance of integrating official narratives with UGC to strengthen OUV communication, foster broader public engagement, and support the sustainable development of WHSs. Full article

Semantic technologies based on ontologies and knowledge graphs are increasingly recognized for their potential to enhance safety, risk, and emergency management in railway systems. This paper presents a systematic literature review aimed at identifying how ontologies, knowledge graphs, and the technologies based on them are applied within the domain of railway safety and assessing their contributions. A total of 53 relevant papers were analyzed using a structured review process, covering four main areas: risk management, safety management, emergency management, and accident analysis. The results reveal that ontologies and knowledge graphs support proactive hazard identification, formalization of safety knowledge, intelligent emergency response, and detailed accident causation modeling. Moreover, they enable semantic interoperability, reasoning, and automation across complex socio-technical railway systems. Despite their benefits, challenges remain regarding data heterogeneity, scalability, and the lack of semantic standardization. This study identifies the most relevant models and technologies, such as SRAC, SRI-Onto, and transformer-based graph neural networks, highlighting their role in advancing intelligent railway safety solutions. This work contributes a detailed map of the current state of semantic applications in railway safety and offers insight into emerging opportunities for future development. Full article

High-resolution remote sensing (HRRS) imagery enables the extraction of cropland information with high levels of detail, especially when combined with the impressive performance of deep convolutional neural networks (DCNNs) in understanding these images. Comprehending the factors influencing DCNNs’ performance in HRRS cropland extraction is of considerable importance for practical agricultural monitoring applications. This study investigates the impact of classifier selection and different training data characteristics on the HRRS cropland classification outcomes. Specifically, Gaofen-1 composite images with 2 m spatial resolution are employed for HRRS cropland extraction, and two county-wide regions with distinct agricultural landscapes in Shandong Province, China, are selected as the study areas. The performance of two deep learning (DL) algorithms (UNet and DeepLabv3+) and a traditional classification algorithm, Object-Based Image Analysis with Random Forest (OBIA-RF), is compared. Additionally, the effects of different band combinations, crop growth stages, and class mislabeling on the classification accuracy are evaluated. The results demonstrated that the UNet and DeepLabv3+ models outperformed OBIA-RF in both simple and complex agricultural landscapes, and were insensitive to the changes in band combinations, indicating their ability to learn abstract features and contextual semantic information for HRRS cropland extraction. Moreover, compared with the DL models, OBIA-RF was more sensitive to changes in the temporal characteristics. The performance of all three models was unaffected when the mislabeling error ratio remained below 5%. Beyond this threshold, the performance of all models decreased, with UNet and DeepLabv3+ showing similar performance decline trends and OBIA-RF suffering a more drastic reduction. Furthermore, the DL models exhibited relatively low sensitivity to the patch size of sample blocks and data augmentation. These findings can facilitate the design of operational implementations for practical applications. Full article

High-precision semantic segmentation of remote sensing imagery is crucial in geospatial analysis. It plays an immeasurable role in fields such as urban governance, environmental monitoring, and natural resource management. However, when confronted with complex objects (such as winding roads and dispersed buildings), existing semantic segmentation methods still suffer from inadequate target recognition capabilities and multi-scale representation issues. This paper proposes a neural network model, LMVMamba (LoRA Multi-scale Vision Mamba), for semantic segmentation of remote sensing images. This model integrates the advantages of convolutional neural networks (CNNs), Transformers, and state-space models (Mamba) with a multi-scale feature fusion strategy. It simultaneously captures global contextual information and fine-grained local features. Specifically, in the encoder stage, the ResT Transformer serves as the backbone network, employing a LoRA fine-tuning strategy to effectively enhance model accuracy by training only the introduced low-rank matrix pairs. The extracted features are then passed to the decoder, where a U-shaped Mamba decoder is designed. In this stage, a Multi-Scale Post-processing Block (MPB) is introduced, consisting of depthwise separable convolutions and residual concatenation. This block effectively extracts multi-scale features and enhances local detail extraction after the VSS block. Additionally, a Local Enhancement and Fusion Attention Module (LAS) is added at the end of each decoder block. LAS integrates the SimAM attention mechanism, further enhancing the model’s multi-scale feature fusion capability and local detail segmentation capability. Through extensive comparative experiments, it was found that LMVMamba achieves superior performance on the OpenEarthMap dataset (mIoU 52.3%, OA 69.8%, mF1: 68.0%) and LoveDA (mIoU 67.9%, OA 80.3%, mF1: 80.5%) datasets. Ablation experiments validated the effectiveness of each module. The final results indicate that this model is highly suitable for high-precision land-cover classification tasks in remote sensing imagery. LMVMamba provides an effective solution for precise semantic segmentation of high-resolution remote sensing imagery. Full article

The escalating prevalence of malware poses a significant threat to digital infrastructure, demanding robust yet efficient detection methods. In this study, we evaluate multiple Convolutional Neural Network (CNN) architectures, including basic CNN, LeNet, AlexNet, GoogLeNet, and DenseNet, on a dataset of 11,000 malware images spanning 452 families. Our experiments demonstrate that CNN models can achieve reliable classification performance across both multiclass and binary tasks. However, we also uncover a critical weakness in that even minimal image perturbations, such as pixel modification lower than 1% of the total image pixels, drastically degrade accuracy and reveal CNNs’ fragility in adversarial settings. A key contribution of this work is spatial analysis of malware images, revealing that discriminative features concentrate disproportionately in the bottom-left quadrant. This spatial bias likely reflects semantic structure, as malware payload information often resides near the end of binary files when rasterized. Notably, models trained in this region outperform those trained in other sections, underscoring the importance of spatial awareness in malware classification. Taken together, our results reveal that CNN-based malware classifiers are simultaneously effective and vulnerable to learning strong representations but sensitive to both subtle perturbations and positional bias. These findings highlight the need for future detection systems that integrate robustness to noise with resilience against spatial distortions to ensure reliability in real-world adversarial environments. Full article

As cities confront rising populations and mounting environmental pressures, waste is rapidly transforming from a logistical liability into a strategic economic resource. In this article, we investigate the evolving nexus between waste and urban economic systems by analyzing over 2000 scientific publications sourced from Web of Science and Scopus. Using advanced semantic embedding and network analysis, we identify seven major research communities at the intersection of digital innovation, circular economy, and smart urban infrastructure. Through PageRank-based influence mapping, we highlight key contributions that shape each thematic cluster—ranging from AI-powered waste classification to blockchain-enabled traceability and IoT-driven logistics. Our results reveal a dynamic and interdisciplinary research landscape where waste valorisation is not only a sustainability imperative but also a driver of urban economic renewal. This study offers both a conceptual map and a methodological framework for understanding how cities can embed intelligence, efficiency, and circularity into waste systems as part of a broader transition to regenerative, data-informed urban economies. Full article

In higher education, institutional quality is traditionally assessed through metrics such as academic programs, research output, educational resources, and community services. However, it is important that their activities align with student expectations, particularly in relation to interactive learning environments, learning management system interaction, curricular and co-curricular activities, accessibility, support services and other learning resources that ensure academic success and, jointly, career readiness. The growing popularity of student engagement metrics as one of the key measures to evaluate institutional efficacy is now a feature across higher education. By monitoring student engagement, institutions assess the impact of existing resources and make necessary improvements or interventions to ensure student success. This study presents a comprehensive analysis of student feedback from the StudentSurvey.ie dataset (2016–2022), which consists of approximately 275,000 student responses, focusing on student self-perception of engagement in the learning process. By using classical topic modelling techniques such as Latent Dirichlet Allocation (LDA) and Bi-term Topic Modelling (BTM), along with the advanced transformer-based BERTopic model, we identify key themes in student responses that can impact institutional strength performance metrics. BTM proved more effective than LDA for short text analysis, whereas BERTopic offered greater semantic coherence and uncovered hidden themes using deep learning embeddings. Moreover, a custom Named Entity Recognition (NER) model successfully extracted entities such as university personnel, digital tools, and educational resources, with improved performance as the training data size increased. To enable students to offer actionable feedback, suggesting areas of improvement, an n-gram and bigram network analysis was used to focus on common modifiers such as “more” and “better” and trends across student groups. This study introduces a fully automated, scalable pipeline that integrates topic modelling, NER, and n-gram analysis to interpret student feedback, offering reportable insights and supporting structured enhancements to the student learning experience. Full article

Effective and fair governance of protected areas (PAs) is essential for their ecological and social sustainability, particularly in contexts of high biodiversity and sociopolitical tensions. This study assessed the governance system of the Serranía del Perijá Regional Natural Park (SPRNP) in Colombia using criteria of effectiveness and justice, through a qualitative methodology grounded in thematic analysis. The research was based on semi-structured interviews and a focus group, with intentional coding supported by artificial intelligence using ATLAS.ti 25 software, which enhanced efficiency and pattern recognition in the construction of a semantic network. This AI-assisted coding approach represents an innovative methodological contribution to the qualitative assessment of PA governance. The findings highlight centralized governance, weak community participation, limited institutional presence, and power asymmetries that undermine equity in decision-making. The exclusion of the Yukpa people from the PA declaration process illustrates broader challenges of Indigenous recognition in Latin American governance contexts. Based on these findings, the study proposes three prospective governance scenarios—community-centered, inter-institutional coordination, and public–private articulation—which offer practical pathways for transforming governance. The study concludes that achieving more equitable and inclusive governance requires institutional strengthening, power redistribution, and the recognition of local knowledge. A viable solution may emerge from an adaptive combination of the proposed scenarios. Full article

Architectural heritage, as a highly symbolized medium of cultural expression, plays a vital role in transmitting collective memory and shaping intercultural tourism experiences. Yet, how visitors from diverse cultural backgrounds perceive and emotionally respond to Chinese architectural symbols remains insufficiently understood. This study addresses this gap by integrating architectural semiotics with cross-cultural psychology to examine perceptual differences across three visitor groups—Mainland China and Hong Kong/Macau/Taiwan (C), East and Southeast Asia (A), and Europe/North America (UA)—at eleven representative Chinese heritage sites. Drawing on 235 in-depth interviews and 1500 online reviews, a mixed-methods design was employed, combining semantic network analysis, grounded theory coding, and affective clustering. The findings reveal that memory structures and cultural contexts shape symbolic perception, that cultural dimensions and affective orientations drive divergent emotional responses, and that interpretive pathways of architectural symbols vary systematically across groups. Specifically, Group C emphasizes collective memory and identity, and Group A engages through structural analogies and regional resonance, while Group UA favors aesthetic form and immersive experiences. These insights inform culturally adaptive strategies for heritage presentation, including memory-anchored curation, comparative cross-regional interpretation, and immersive digital storytelling. By advancing a micro-level model of “architectural symbol–perceptual theme–emotional response–perceptual mechanism,” this research not only enriches theoretical debates on cross-cultural heritage perception but also offers practical guidance for inclusive and resonant heritage interpretation in a global tourism context. Full article

The widespread emergence of multimodal data on social platforms has presented new opportunities for sentiment analysis. However, previous studies have often overlooked the issue of detail loss during modal interaction fusion. They also exhibit limitations in addressing semantic alignment challenges and the sensitivity of modalities to noise. To enhance analytical accuracy, a novel model named MAHFNet is proposed. The proposed architecture is composed of three main components. Firstly, an attention-guided gated interaction alignment module is developed for modeling the semantic interaction between text and image using a gated network and a cross-modal attention mechanism. Next, a contrastive learning mechanism is introduced to encourage the aggregation of semantically aligned image-text pairs. Subsequently, an intra-modality emotion extraction module is designed to extract local emotional features within each modality. This module serves to compensate for detail loss during interaction fusion. The intra-modal local emotion features and cross-modal interaction features are then fed into a hierarchical gated fusion module, where the local features are fused through a cross-gated mechanism to dynamically adjust the contribution of each modality while suppressing modality-specific noise. Then, the fusion results and cross-modal interaction features are further fused using a multi-scale attention gating module to capture hierarchical dependencies between local and global emotional information, thereby enhancing the model’s ability to perceive and integrate emotional cues across multiple semantic levels. Finally, extensive experiments have been conducted on three public multimodal sentiment datasets, with results demonstrating that the proposed model outperforms existing methods across multiple evaluation metrics. Specifically, on the TumEmo dataset, our model achieves improvements of 2.55% in ACC and 2.63% in F1 score compared to the second-best method. On the HFM dataset, these gains reach 0.56% in ACC and 0.9% in F1 score, respectively. On the MVSA-S dataset, these gains reach 0.03% in ACC and 1.26% in F1 score. These findings collectively validate the overall effectiveness of the proposed model. Full article

Emerging complex multi-step attacks such as Advanced Persistent Threats (APTs) pose significant risks to national economic development, security, and social stability. Effectively detecting these sophisticated threats is a critical challenge. While deep learning methods show promise in identifying unknown malicious behaviors, they often struggle with fragmented modal information, limited feature representation, and generalization. To address these limitations, we propose GATransformer, a new dual-modal detection method that integrates topological structure analysis with temporal sequence modeling. Its core lies in a cross-attention semantic fusion mechanism, which deeply integrates heterogeneous features and effectively mitigates the constraints of unimodal representations. GATransformer reconstructs network behavior representation via a parallel processing framework in which graph attention captures intricate spatial dependencies, and self-attention focuses on modeling long-range temporal correlations. Experimental results on the CIDDS-001 and CIDDS-002 datasets demonstrate the superior performance of our method compared to baseline methods with detection accuracies of 99.74% (nodes) and 88.28% (edges) on CIDDS-001 and 99.99% and 99.98% on CIDDS-002, respectively. Full article

Dynamic path generation in complex transportation networks is essential for intelligent transportation systems. Traditional methods, such as shortest path algorithms or heuristic-based models, often fail to capture real-world travel behaviors due to their reliance on simplified assumptions and limited ability to handle long-range dependencies or non-linear patterns. To address these limitations, we propose PathGen-LLM, a large language model (LLM) designed to learn spatial–temporal patterns from historical paths without requiring handcrafted features or graph-specific architectures. Exploiting the structural similarity between path sequences and natural language, PathGen-LLM converts spatiotemporal trajectories into text-formatted token sequences by encoding node IDs and timestamps. This enables the model to learn global dependencies and semantic relationships through self-supervised pretraining. The model integrates a hierarchical Transformer architecture with dynamic constraint decoding, which synchronizes spatial node transitions with temporal timestamps to ensure physically valid paths in large-scale road networks. Experimental results on real-world urban datasets demonstrate that PathGen-LLM outperforms baseline methods, particularly in long-distance path generation. By bridging sequence modeling and complex network analysis, PathGen-LLM offers a novel framework for intelligent transportation systems, highlighting the potential of LLMs to address challenges in large-scale, real-time network tasks. Full article