Search Results (1,817)

This paper examines whether and how analyst logical inconsistency affects the stock price crash risk in China’s A-share market. The sample includes 3784 listed non-financial companies. The final sample comprises 12,460 firm-year observations from 2016 to 2023. Specifically, we employ the open source Qwen1.5-14B-Chat model, an instruction-tuned generative large language model, to measure analyst logical inconsistency, classify the sentiment expressed in analyst reports, and determine the differences between earnings forecasts and textual tone. Using a panel fixed-effect model for basic regression, and applying two-stage least squares and propensity score matching to deal with endogenous problems, we find that analyst logical inconsistency significantly increases the stock price crash risk. The results remain robust to alternative variable definitions, additional control variables, alternative sample periods, and more stringent fixed-effects specifications. The mechanism test shows that the analyst logical inconsistency increases the stock price crash risk through three channels: increased financial risk, reduced investment efficiency, and degraded information disclosure quality. Heterogeneity analysis also shows that this positive impact is strongest in companies with high media coverage, good corporate reputation and low ESG performance. Our research results are helpful to the study of information intermediaries and stock price crash risk by introducing measures for the quality of analyst reports based on large language models, and provide useful suggestions for regulators and investors in emerging markets. Full article

The rapid rise in autonomous AI agents is reshaping personal task automation and human–AI collaboration. This study develops a theoretically informed framework, drawing on the Elaboration Likelihood Model, the Technology Acceptance Model, and the sense of virtual community, to examine how comment topics and sentiment polarity relate to comment endorsement and how these associations vary across levels of reply activity. Using 55,502 raw comments from 1000 popular YouTube videos about OpenClaw, we retained 14,974 high-quality English comments for analysis. We used BERTopic to identify the main comment topics, and RoBERTa to classify sentiment polarity. The results show that (1) topics related to functionality are positively associated with comment endorsement, whereas those related to security and hardware are negatively associated with comment endorsement, and (2) the interaction associations involving reply count vary across different topics and sentiment polarity. These findings reveal users’ key concerns and emotional responses toward the emerging OpenClaw technology and offer theoretically informed and practical implications for iteratively improving AI agent tools. Full article

Cultural ecosystem services (CES) are widely measured but rarely attributed to the specific landscape elements that generate them. We coupled element-level image recognition with service-level text analysis across 20,447 reviews and 18,997 photographs from three social media platforms covering 15 national-, provincial-, and municipal-level forest parks in Guangzhou, China (2020–2024). A 2660-term dictionary quantified nine CES; a fine-tuned deep learning classifier identified 28 landscape elements; multiple correspondence analysis described the element-service structure and ridge regression tested associations within four case parks. Natural elements aligned with ecological, aesthetic, and inspirational services and built elements with cultural, social, and wellness services, but two results qualify that division. First, individual elements were associated with different services in opposite directions: seasonal forest with higher aesthetic appreciation but lower recreation, water features with higher aesthetic value but lower inspiration. Second, the mapping differed across the four case parks rather than holding constant; in the municipal case park, only built facilities positively predicted social interaction. These four parks are illustrative exemplars rather than representative samples of their tiers. Dynamic biological elements (birds 0.43%, insects 0.12%, fish 0.08%) ranked lowest of all 28 elements. These are descriptive associations awaiting mechanistic explanation; on that reading, planning for CES would be better organized around intended services than element inventories, calibrated to the individual park. Full article

Federated learning (FL) has emerged as an important distributed machine learning paradigm allowing many users to train a model together without sharing raw data. However, FL’s decentralized design makes it very susceptible to both Byzantine and coordinated model poisoning attacks, where a few malicious rounds of gradients are strategically inserted to reduce the overall integrity of the model. Even the existing Byzantine-resilient aggregation methods such as geometric median, Krum, and trimmed mean are vulnerable to sophisticated, coordinated poisoning attacks that leverage statistical gaps in outlier detection. In this article, we present FedSentinel, a novel Byzantine-resilient federated learning framework that combines cryptographic gradient attestation with adaptive trust-weighted aggregation to protect against coordinated model-poisoning attacks, which are among the most serious challenges. Three key innovations are introduced in FedSentinel Cryptographic Gradient Attestation Protocol (CGAP) that utilizes commitment schemes and zero-knowledge range proofs to ensure the integrity of the gradients and verify that submitted updates satisfy predefined norm and direction constraints; in the current protocol configuration, CGAP provides gradient integrity verification rather than full gradient confidentiality from the server Dynamic Trust-Weighted Robust Aggregation (DT-RoA): the computation and updating of per-client trust scores based on historical gradient consistency and cross-validation signals; and Coordinated Attack Detection Engine (CADE): based on spectral analysis of the gradient covariance matrices, it detects and isolates colluding Byzantine clients. By participating in up to 100 clients on CIFAR-10, CIFAR-100, FEMNIST, and Sentiment140 datasets, extensive experiments conducted under up to 30% Byzantine adversaries reveal that FedSentinel achieves 91.36% average global accuracy, whereas state-of-the-art defenses such as FLTrust, FLAME, RoFL, ShieldFL, and DnC achieve 83.83–86.94%. FedSentinel outperforms these defenses by 4.42–7.53% in terms of accuracy, while decreasing attack success rates by 53.2% under coordinated Byzantine backdoor attacks. The proposed framework offers a promising approach for federated learning that is verifiable and trustworthy in adversarial environments. Full article

The news media in Gulf countries are among the most important institutions shaping the public’s understanding of climate action. This study aims to explore the longitudinal developments, thematic framing, and sentiment tones of Omani English-language newspapers’ sustainability discourse during the period 2015–2025. The study applied a multi-stage computational method to analyze a deduplicated corpus of 709 news articles from two English-language dailies of Oman: Times of Oman and Oman Daily Observer for a time period from 2015 to 2025. Latent Dirichlet Allocation (LDA) topic modeling was used to group topics into five types of macro frames. Results show that the focus of the Omani media is very localized, and the dominant frames are “Corporate ESG, Business & Green Investment” and “Clean Energy Policy, Research & Institutional Governance”. Furthermore, a sentiment analysis of headlines using VADER showed clearly defined sentiment polarity; domestic economic transition headlines were characterized by high positive polarity, while global climate crisis headlines had a higher negative polarity. The study concludes that the press in Oman is in a strategic process of “nationalizing” the climate change issue, moving from the conventional disaster discourse to making sustainability a local opportunity for economic diversification, technological modernization, and national resilience. Full article

Social media has become an important tool for understanding visitor experiences in museums and cultural heritage sites. This literature review identifies, organizes, and thematically synthesizes existing studies on museum visitor experience based on social media data. It examines 41 studies retrieved from Scopus and Web of Science and published between 2017 and 2026. Furthermore, the review examines the selected studies across six dimensions: social media platforms, types of user-generated data, the role of digital interactions, the museums and cultural heritage sites studied, the analytical methodologies applied, and the main findings on visitor experience. The findings indicate that TripAdvisor is the most frequently used platform for collecting textual reviews and star ratings, whereas Instagram and Flickr are mainly used for visual and spatial data. Most studies rely on computational methods, often combined with quantitative techniques, while qualitative approaches are used less frequently. The identified methods include content analysis, statistical analysis, sentiment analysis, topic modeling, machine learning, image analysis, and spatial analysis. Across the reviewed studies, visitor experience is examined as a multidimensional phenomenon encompassing emotions, service quality, authenticity, historical connection, aesthetics, education, and social participation. Finally, the review identifies recurring themes across the dimensions and synthesizes them into broader research streams. These are brought together in an integrative synthesis framework that organizes existing research, highlights research gaps, and outlines directions for future studies. Full article

The rapid growth of organisational textual data has increased the value of Natural Language Processing (NLP) and semantic analytics for strategic decision support. However, many employees still lack the knowledge and skills needed to evaluate AI-generated information critically. This study develops an explainable Management Information Systems (MIS) framework that integrates NLP-based semantic analytics with PLS-SEM to examine the relationship between AI illiteracy and strategic decision quality. A convergent mixed-methods design with sequential analytical integration was used with a valid sample of 200 knowledge workers from public organisations in Saudi Arabia across six industries. The sample included employees from Saudi Arabia, Egypt, Jordan, Sudan, Syria, India, and the Philippines. Quantitative data were analysed using PLS-SEM, while textual data were analysed using Sentence-BERT, BERTopic, semantic network analysis, and Aspect-Based Sentiment Analysis. The results showed that higher AI illiteracy was negatively associated with strategic decision quality and positively associated with automation bias, uncritical trust in AI, and cognitive offloading. Digital proficiency and AI governance awareness weakened the negative association between AI illiteracy and decision quality, while functional-background differences were examined through multigroup analysis. The semantic analysis identified six themes: AI competency, decision trust, AI governance, decision support, organisational learning, and risk awareness. Sentiment analysis showed positive views of productivity and decision support, together with concerns about algorithmic bias, explainability, transparency, and AI governance. The study contributes an integrated human–AI decision vulnerability framework in which semantic evidence complements structural modelling and provides a clearer understanding of AI-related competency, reliance, governance, and decision-support issues. Full article

Short-term financial risk in intelligent trading systems is reflected not only in prices, trading volumes, and textual sentiment but also in infrastructure operating states, including server workload, device power consumption, network latency, and packet loss rate. We propose HSF-LLMNet, a hardware sensor and large language model semantic fusion network for jointly modeling external information shocks and infrastructure responses. A large language model extracts event category, sentiment polarity, risk intensity, and semantic uncertainty from financial texts. Reliability-aware temporal modeling handles sensor missingness, drift, and abnormal noise, while asynchronous soft alignment, bidirectional cross-attention, and reliability-aware gated fusion integrate irregular textual events with continuous hardware signals. The model jointly predicts market direction, realized volatility, and three-level risk over the subsequent 30 min. Experiments were conducted on eight Chinese A-share indices: the SSE Composite Index (000001.SH), SSE 50 Index (000016.SH), CSI 300 Index (000300.SH), STAR 50 Index (000688.SH), CSI 500 Index (000905.SH), CSI 1000 Index (000852.SH), Shenzhen Component Index (399001.SZ), and ChiNext Index (399006.SZ). The common observation period for market, textual, and hardware data extended from 1 March 2024 to 30 June 2025. After data cleaning, timestamp matching, and multimodal temporal alignment, 169,208 aligned asset–time prediction windows were retained for the 30 min forecasting task. Realized volatility was defined as the square root of the sum of squared one-minute log returns over the future 30 min interval. The three-level risk label was constructed from future realized volatility, absolute 30 min return, and liquidity stress, with all thresholds estimated exclusively from the training portion of each fold. A sample was labeled high risk when at least two of the three indicators exceeded their 85th-percentile thresholds or when any indicator exceeded its 95th-percentile threshold. It was labeled medium risk when, after excluding high-risk samples, at least two indicators exceeded their 60th-percentile thresholds or any indicator exceeded its 85th-percentile threshold; all remaining samples were labeled low risk. Results showed that HSF-LLMNet achieved an accuracy of 78.62%, a precision of 78.14%, a recall of 77.83%, an F1-score of 77.98%, an area under the receiver operating characteristic curve of 84.91%, and a Matthews correlation coefficient of 57.36% for directional prediction. For realized-volatility regression, the MAE, RMSE, MAPE, and $R^2$ were 0.0089, 0.0135, 9.21%, and 0.812, respectively. For high-risk-event warning, the mean effective warning time, defined as the interval between the first valid alarm and the corresponding event, was 15.37 min; the false-alarm rate and missed-alarm rate were 6.82% and 8.14%, respectively. Ablation experiments showed performance reductions after removing semantic encoding, sensor-reliability estimation, asynchronous alignment, bidirectional cross-attention, gated fusion, or multi-task learning. These results indicate that textual events and infrastructure operating states provide complementary information for quantitative risk analytics and fintech applications. Full article

The pervasive rise in digital platforms has fundamentally recalibrated the relationship between urban communities and their cultural landscapes, reconfiguring not only the sensory experience of these environments but also their interpretive frames and the everyday contestations that surround them. Using coffee spaces in Beijing as an empirical lens, this study integrates 630 validated POI locations, 287,632 online reviews from Dianping, GIS spatial analysis, LDA topic modelling, semiotic coding, and a fine-tuned RoBERTa (Robustly Optimized BERT Pretraining Approach) text classifier to examine the convergence between platform-mediated public perceptions and urban spatial differentiation. Our spatial interrogation identifies a statistically significant centre–periphery divide within this platform-curated corpus—a pattern that describes the geography of platform-mediated visibility rather than the total physical distribution of coffee outlets in Beijing, characterised by pronounced high-density agglomerations nesting within historic urban cores, central business districts, nascent knowledge hubs, and gentrifying lifestyle enclaves. Three landscape prototypes are derived from the data—business corridors, cultural experimental fields, and social-media spectacles. Each type exhibits discernibly different patterns of spatial density, adjacent land-use configurations, perceived experiential qualities, and underlying symbolic discourses. While positive sentiments dominate platform discourse, localised anxieties emerge regarding visual homogenization, intense queuing, premium pricing, and loss of place authenticity. The 630 venues analysed here represent a platform-visibility stratum—establishments sufficiently reviewed to feature on digital intermediaries—and the spatial patterns we map describe the geography of this visibility rather than the total physical distribution of coffee outlets. Full article

This paper focuses on China’s gold futures market. Using East Money investor posts and the FinBERT2 model, we construct a multidimensional sentiment indicator comprising overall sentiment, positive/negative intensity, and six fine-grained sentiment categories (happiness, sadness, fear, anger, disgust, and neutral). With high-frequency 5 min data, we compute realized volatility and, within the HAR-RV framework, systematically examine the in-sample and out-of-sample predictive power, asymmetry, and inter-period heterogeneity of sentiment dimensions. The results show that investor sentiment significantly and robustly predicts volatility, with gains increasing over horizons, relying on multi-scale cumulative effects. Predictions are asymmetric: negative sentiment drives volatility while positive sentiment does not. Among fine-grained sentiments, happiness and anger are strongest; fear and sadness are ineffective; and disgust has an effect only in long-term routine forecasts but fails under extreme volatility. During the Russia–Ukraine conflict, sadness replaces happiness and anger as the dominant predictor (long-term MSE: 0.01579 vs. benchmark 0.04384). In trending bull markets, the predictive power of happiness and positive/negative intensity is amplified (full-sample R² gains: 2.70% and 2.89%, vs. 17.08% and 9.69% in bull periods). This study reveals the multidimensional, asymmetric effects and intertemporal heterogeneity of sentiment on forecasts of gold futures volatility, providing a theoretical and empirical foundation for regime-adaptive early-warning systems and risk management. Full article

Multimodal sentiment analysis aims to integrate heterogeneous textual, visual, and acoustic information for effective emotion understanding. However, existing methods often suffer from insufficient cross-modal interaction modeling, limited adaptability in multimodal fusion, and inadequate suppression of modality-specific noise under complex conversational scenarios. To address these challenges, this paper proposes a framework for learning adaptive cross-modal interactions for multimodal sentiment analysis. The proposed framework consists of three stages: modality-aware preprocessing, heterogeneous representation learning, and adaptive multimodal fusion. First, a unified preprocessing strategy is designed to improve cross-modal consistency through textual normalization, speaker-aware visual alignment, and utterance-level acoustic representation enhancement. Second, modality-specific encoders are constructed to capture complementary semantic, spatial, and utterance-level acoustic characteristics from textual, visual, and acoustic modalities, respectively. Third, an adaptive fusion framework is introduced to explicitly model cross-modal interactions, dynamically estimate the importance of different modality combinations, and further calibrate discriminative feature channels through channel attention. By jointly performing modality-level interaction learning and channel-wise feature refinement, the proposed framework effectively enhances multimodal representation capability for sentiment classification. Extensive experiments conducted on the CMU-MOSI and MELD benchmark datasets demonstrate that our framework consistently outperforms previous methods. In particular, the proposed model achieves 90.27% accuracy and 90.26% F1-score on CMU-MOSI, together with 66.57% accuracy and 66.21% F1-score on MELD. Additional ablation studies and qualitative analyses further validate the effectiveness of the proposed preprocessing strategy, modality-specific representation learning, and adaptive fusion mechanism. Full article

Pre-trained language models are typically adapted to downstream tasks via full fine tuning. However, this entails substantial computational and memory overhead. Parameter-efficient alternatives such as Low-Rank Adaptation (LoRA), Quantized Low-Rank Adaptation (QLoRA), and frozen backbone mitigate these costs, but empirical evidence on how their performance–cost trade-offs change under low labeled data in Arabic remains limited. This paper compares four adaptation strategies: full fine tuning, frozen backbone, LoRA, and QLoRA for Arabic binary sentiment classification on the Hotel Arabic Reviews Dataset, using CAMeLBERT-Mix as the pre-trained encoder. The methods are evaluated under a unified experimental setting at three labeled-data levels: the full training set, 100 samples per class, and 25 samples per class. The evaluation metrics are reported as means and standard deviations across five random seeds. At the full-data level, full fine tuning, LoRA, and QLoRA achieve macro-F1 scores between 0.9569 and 0.9579 and are comparable within seed variability, while the frozen backbone exhibits performance that is approximately ten points lower. LoRA and QLoRA use approximately 35.0% less peak GPU memory than full fine tuning but require longer training times. Under reduced-data conditions, full fine tuning outperforms all other adaptation strategies with the differences being statically significant. Full article

Aspect-based Sentiment Analysis (ABSA) is a fine-grained sentiment classification task that aims to predict the sentiment polarity associated with aspect terms in sentences. Traditional methods based on syntactic and semantic dependency trees are insufficient for capturing contextual sentence features. To address this, we propose a Knowledge-Enhanced and Multidimensional Feature Fusion (KEMFF) model for ABSA, which captures sentiment feature representations across multiple dimensions, including syntax, semantics, and knowledge. First, the pre-trained model RoBERTa is used to obtain embeddings of sentences and aspect terms. Then, a syntactic dependency parser and a graph convolutional network are utilized to learn syntactic features. Meanwhile, an Abstract Meaning Representation (AMR)-based parser is employed to construct semantic relations, and axial attention is used to aggregate incoming and outgoing semantic dependencies. Furthermore, external knowledge is embedded, and an attention mechanism is employed to obtain aspect-specific knowledge representations, thereby complementing syntactic and semantic representations with external lexical knowledge. Finally, multidimensional features are fused and passed to a softmax classifier for predicting sentiment polarities. Unlike previous models that mainly focus on either syntax–semantic fusion or knowledge-enhanced graph propagation, KEMFF explicitly models syntax, semantics, and lexical knowledge in three parallel branches and aligns them into a unified aspect-level representation. Experiments on Laptop14, Restaurant14, and Twitter datasets show that KEMFF achieves the best performance among the compared baselines on Laptop14 and Restaurant14, and it obtains competitive results on Twitter. Full article

With social media as the main channel for college students to express emotions, sentiment public opinion analysis in big data environments poses three core challenges to campus sentiment monitoring and psychological counseling: severe data noise interference, insufficient multi-attribute feature extraction, and the trade-off between recognition accuracy and inference efficiency. This paper proposes a university student public opinion prediction model integrating multi-attribute decision-making and BERT–Mamba. First, an anti-interference matching filter cleans raw data by filtering out advertisements and irrelevant comments to improve data quality. Second, a multi-attribute decision object model extracts quantifiable attributes covering media sources, themes, and temporal dimensions. Third, BERT generates textual sentiment representations, and a three-stage deep feature extraction architecture with Mamba balances accuracy and efficiency. Finally, multi-attribute features and sentiment representations are fused for dynamic public opinion prediction. Validated using the ChnSentiCorp Chinese sentiment analysis benchmark dataset and university student Weibo public opinion corpus, the model achieves 97.44% average sentiment recognition accuracy. It provides technical support for universities to understand student sentiment trends and address negative public opinions, with practical value for enhancing campus public opinion monitoring and assisting mental health counseling. Full article

Social media is a vital platform for policy communication, yet research rarely integrates content and structural features to evaluate communication effects multidimensionally. This study proposes a convergent framework combining text mining and social network analysis to assess policy communication effects and underlying mechanisms. Using China’s upgraded Mass Entrepreneurship and Innovation policy on Sina Weibo as a case, we analyze communication breadth, depth, audience sentiment, thematic focus, network topology, key nodes, and community characteristics. The results reveal that (1) communication breadth is dominated by official communicators, while audiences drive interactive depth, reflecting a “centralized broadcasting, decentralized engagement” model; (2) influential users express more positive attitudes than ordinary audiences; (3) discussions diversify from core innovation themes to micro-level concerns like regional development and talent policies; (4) the network shows loose global structure but strong local clustering, with bridging nodes posting less polarized, broader content. Theoretically, this study offers behavioral-level observations that align with key corollaries of the Spiral of Silence Theory—the tendency for individuals with deviating views to shift toward lower-visibility participation. These pattern-level findings offer a complementary empirical perspective on opinion expression in digital policy contexts. Practically, the findings offer preliminary insights that may inform adaptive, decentralized strategies for enhancing policy diffusion in similar social media contexts. Full article