Search Results (81)

To address the challenges of extensive domain expertise and deficient semantic comprehension in the digital preservation of ancient ceramics, this paper proposes a knowledge question answering (QA) system integrating Low-Rank Adaptation (LoRA) fine-tuning and Graph Retrieval-Augmented Generation (GraphRAG). First, textual information of ceramic images is generated using the GLM-4V-9B model. These texts are then enriched with domain literature to produce ancient ceramic QA pairs via ERNIE 4.0 Turbo, culminating in a high-quality dataset of 2143 curated question–answer groups after manual refinement. Second, LoRA fine-tuning was employed on the Qwen2.5-7B-Instruct foundation model, significantly enhancing its question-answering proficiency specifically for the ancient ceramics domain. Finally, the GraphRAG framework is integrated, combining the fine-tuned large language model with knowledge graph path analysis to augment multi-hop reasoning capabilities for complex queries. Experimental results demonstrate performance improvements of 24.08% in ROUGE-1, 34.75% in ROUGE-2, 29.78% in ROUGE-L, and 4.52% in BERTScore_F1 over the baseline model. This evidence shows that the synergistic implementation of LoRA fine-tuning and GraphRAG delivers significant performance enhancements for ceramic knowledge systems, establishing a replicable technical framework for intelligent cultural heritage knowledge services. 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

Efficient emergency load transfer is crucial for ensuring the power system’s safe operation and reliable power supply. However, traditional load transfer methods that rely on human experience have limitations, such as slow response times and low efficiency, which make it difficult to address complex and diverse fault scenarios effectively. Therefore, this paper proposes an emergency load transfer method based on knowledge graphs to achieve intelligent management and efficient retrieval of emergency knowledge. Firstly, a named entity recognition model based on ERNIE-BiGRU-CRF is constructed to automatically extract key entities and relationships from the load transfer plan texts, obtaining information such as fault names, fault causes, and operation steps. Secondly, a power system emergency load transfer knowledge graph is constructed based on the extracted structured knowledge, which is efficiently stored using a graph database and enables the visualization and interactive query of knowledge. Finally, real power system fault cases prove that the proposed method can effectively improve the retrieval efficiency of fault knowledge and provide intelligent support for online emergency load transfer decisions. Full article

With the accelerated digital transformation in education, the efficient integration of massive multimodal instructional resources and the support for interactive question answering (QA) remains a prominent challenge. This study introduces Multimodal Disciplinary Knowledge-Augmented Generation (MDKAG), a framework integrating retrieval-augmented generation (RAG) with a multimodal disciplinary knowledge graph (MDKG). MDKAG first extracts high-precision entities from digital textbooks, lecture slides, and classroom videos by using the Enhanced Representation through Knowledge Integration 3.0 (ERNIE 3.0) model and then links them into a graph that supports fine-grained retrieval. At inference time, the framework retrieves graph-adjacent passages, integrates multimodal data, and feeds them into a large language model (LLM) to generate context-aligned answers. An answer-verification module checks semantic overlap and entity coverage to filter hallucinations and triggers incremental graph updates when new concepts appear. Experiments on three university courses show that MDKAG reduces hallucination rates by up to 23% and increases answer accuracy by 11% over text-only RAG and knowledge-augmented generation (KAG) baselines, demonstrating strong adaptability across subject domains. The results indicate that MDKAG offers an effective route for scalable knowledge organization and reliable interactive QA in education. Full article

The microstructure and mechanical properties of welded joints of API 5L X-52 steel plates cladded with AISI 316L-Si austenitic stainless steel were evaluated. The gas metal arc welding process with pulsed arc (GMAW-P) and controlled arc oscillation were used to join the bimetallic plates. After the root welding pass, buttering with an ERNiCrMo-3 filler wire was performed and multi-pass welding followed using an ER70S-6 electrode. The results obtained by optical and scanning electron microscopy indicated that the shielding atmosphere, welding parameters, and electric arc oscillation enabled good arc stability and proper molten metal transfer from the filler wire to the sidewalls of the joint during welding. Vickers microhardness (HV) and tensile tests were performed for correlating microstructural and mechanical properties. The mixture of ERNiCrMo-3 and ER70S-6 filler materials presented fine interlocked grains with a honeycomb network shape of the Ni–Fe mixture with Ni-rich grain boundaries and a cellular-dendritic and equiaxed solidification. Variation of microhardness at the weld metal (WM) in the middle zone of the bimetallic welded joints (BWJ) is associated with the manipulation of the welding parameters, promoting precipitation of carbides in the austenitic matrix and formation of martensite during solidification of the weld pool and cooling of the WM. The BWJ exhibited a mechanical strength of 380 and 520 MPa for the yield stress and ultimate tensile strength, respectively. These values are close to those of the as-received API 5L X-52 steel. Full article

Building Information Modeling (BIM) has excellent potential to enhance building operation and maintenance. However, as a standardized data format in the architecture, engineering, and construction (AEC) industry, the retrieval of BIM information generally requires specialized software. Cumbersome software operations prevent its effective application in the actual operation and management of buildings. This paper presents BIMCoder, a model designed to translate natural language queries into structured query statements compatible with professional BIM software (e.g., BIMserver v1.5). It serves as an intermediary component between users and various BIM platforms, facilitating access for users without specialized BIM knowledge. A dedicated BIM information query dataset was constructed, comprising 1680 natural language query and structured BIM query string pairs, categorized into 12 groups. Three classical pre-trained large language models (LLMs) (ERNIE 3.0, Llama-13B, and SQLCoder) were evaluated on this dataset. A fine-tuned model based on SQLCoder was then trained. Subsequently, a fusion model (BIMCoder) integrating ERNIE and SQLCoder was designed. Test results demonstrate that the proposed BIMCoder model achieves an outstanding accurate matching rate of 87.16% and an Execution Accuracy rate of 88.75% for natural language-based BIM information retrieval. This study confirms the feasibility of natural language-based BIM information retrieval and offers a novel solution to reduce the complexity of BIM system interaction. Full article

The purpose of this study is to compare the accuracy and readability of Coronavirus Disease 2019 (COVID-19)-prevention and control knowledge texts generated by four current generative artificial intelligence (AI) models—two international models (ChatGPT and Gemini) and two domestic models (Kimi and Ernie Bot)—and to evaluate the other performance characteristics of texts generated by domestic and international models. This paper uses the questions and answers in the COVID-19 prevention guidelines issued by the U.S. Centers for Disease Control and Prevention (CDC) as the evaluation criteria. The accuracy, readability, and comprehensibility of the texts generated by each model are scored against the CDC standards. Then the neural network model in the intelligent algorithms is used to identify the factors that affect readability. Then the medical topics of the generated text are analyzed using text analysis technology. Finally, a questionnaire-based manual scoring approach was used to evaluate the AI-generated texts, which was then compared to automated machine scoring. Accuracy: domestic models have higher textual accuracy, while international models have higher reliability. Readability: domestic models produced more fluent and publicly accessible language; international models generated more standardized and formally structured texts with greater consistency. Comprehensibility: domestic models offered superior readability, while international models were more stable in output. Readability factors: the average words per sentence (AWPS) emerged as the most significant factor influencing readability across all models. Topic analysis: ChatGPT emphasized epidemiological knowledge; Gemini focused on general medical and health topics; Kimi provided more multidisciplinary content; and Ernie Bot concentrated on clinical medicine. From the empirical results, it can be found that the manual and machine scoring are highly consistent in the indicators SimHash and FKGL, which proves the effectiveness of the evaluation method proposed in this paper. Conclusion: Texts generated by domestic models are more accessible and better suited for public education, clinical communication, and health consultations. In contrast, the international model has a higher accuracy in generating expertise, especially in epidemiological studies and assessing knowledge literature on disease severity. The inclusion of manual evaluations confirms the reliability of the proposed assessment framework. It is therefore recommended that future AI-generated knowledge systems for infectious disease control balance professional rigor with public comprehensibility, in order to provide reliable and accessible reference materials during major infectious disease outbreaks. Full article

This study employs DFT+U calculations to investigate the ferromagnetic properties of ErAl₂ and ErNi₂ Laves phases under an external hydrostatic pressure P (0 GPa ≤ P ≤ 1.0 GPa). The calculated magnetic moments per formula unit for both crystalline structures align with experimentally reported values: 4.40 μ_B/f.u. in the hard magnetization <001> axis for ErAl₂ and 5.56 μ_B/f.u. in the easy magnetization <001> axis for ErNi₂. The DFT results indicate that the magnetic moment remains unchanged up to 1 GPa of hydrostatic pressure, with no structural instabilities observed, as evidenced by a nearly constant formation energy for ErAl₂ and ErNi₂ alloys. The simulations confirm that the magnetic behavior of ErAl₂ is primarily driven by the electrons localized in the f orbitals. In contrast, for ErNi₂, both d and f orbitals significantly contribute to the total magnetic moment. Finally, the electronic specific heat coefficient was calculated and reported as a function of hydrostatic pressure up to P = 1.0 GPa for each Laves phase. Full article

The inherent uncertainty and information asymmetry in financial markets create significant challenges for accurate price forecasting. Although investor sentiment analysis has gained traction in recent research, the temporal dimension of sentiment dynamics remains underexplored. This study develops a novel framework that enhances stock price prediction by integrating time-partitioned investor sentiment, while improving model interpretability via Shapley additive explanations (SHAP) analysis. Employing the ERNIE (enhanced representation through knowledge integration) 3.0 model for sentiment extraction from China’s Eastmoney Guba stock forum, we quantitatively distinguish intraday and post-market investor sentiment then integrate these temporal components with technical indicators through neural network architecture. Our results indicate that temporal sentiment partitioning effectively reduces uncertainty. Empirical evidence demonstrates that our long short-term memory (LSTM) model integrating intraday and post-market sentiment indicators achieves better prediction accuracy, and SHAP analysis reveals the importance of intraday and post-market investor sentiment to stock price prediction models. Implementing quantitative trading strategies based on these insights generates significantly more annualized returns for representative stocks with controlled risk, outperforming sentiment-agnostic and non-temporal sentiment models. This research provides methodological innovations for processing temporal unstructured data in finance, while the SHAP framework offers regulators and investors actionable insights into sentiment-driven market dynamics. Full article

Effectively identifying factors related to user satisfaction is crucial for evaluating customer experience. This study proposes a two-phase analytical framework that combines natural language processing techniques with hierarchical decision-making methods. In Phase 1, an ERNIE-LSTM-based emotion model (ELEM) is used to detect fake reviews from 4016 smartphone evaluations collected from JD.com (accuracy: 84.77%, recall: 84.86%, F1 score: 84.81%). The filtered genuine reviews are then analyzed using Biterm Topic Modeling (BTM) to extract key satisfaction-related topics, which are weighted based on sentiment scores and organized into a multi-criteria evaluation matrix through the Analytic Hierarchy Process (AHP). These topics are further clustered into five major factors: user-centered design (70.8%), core performance (10.0%), imaging features (8.6%), promotional incentives (7.8%), and industrial design (2.8%). This framework is applied to a comparative analysis of two smartphone stores, revealing that Huawei Mate 60 Pro emphasizes performance, while Redmi Note 11 5G focuses on imaging capabilities. Further clustering of user reviews identifies six distinct user groups, all prioritizing user-centered design and core performance, but showing differences in other preferences. In Phase 2, a comparison of word frequencies between product reviews and community Q and A content highlights hidden user concerns often missed by traditional single-source sentiment analysis, such as screen calibration and pixel density. These findings provide insights into how product design influences satisfaction and offer practical guidance for improving product development and marketing strategies. Full article

With the increasing complexity of modern power systems, traditional risk assessment methods relying on expert experience and historical data face challenges in accuracy and adaptability. This study proposes a knowledge graph-augmented ERNIE-CNN method to enhance risk assessment in secondary power system operations. First, we construct a domain-specific knowledge graph by integrating expert knowledge and operational standards, which enhances semantic understanding and logical reasoning capabilities. Second, an improved ERNIE-CNN model is developed, incorporating an attention mechanism to effectively fuse semantic features and spatial patterns from operational texts. The experimental results on a dataset of 3240 secondary operation records demonstrate the model’s superior performance, achieving precision, recall, and F1-scores of 0.878, 0.861, and 0.869, respectively, outperforming benchmarks like BERT. Furthermore, a visualization of the knowledge graph is implemented, providing interpretable decision support for risk management. The proposed method offers a robust framework for automating risk assessment in power systems, with potential applications in smart grid maintenance and safety-critical operational planning. Full article

The rapid development of generative artificial intelligence (GenAI) technology has triggered extensive discussions about its potential applications in sustainable higher education. Based on the technology acceptance model (TAM) and task–technology fit (TTF) theory, this research aimed to investigate the current situations and challenges of Chinese university students using GenAI in four typical task scenarios. This was performed using a cross-sectional research design. The data were collected via questionnaire, with 486 undergraduates from a Chinese university participating. The data analysis methods include descriptive statistics, inferential statistics, and content analysis. The results show that more than 70% of university students actively use GenAI, but nearly half of them are not very proficient in its use. Doubao and ERNIE Bot are the GenAI tools they prefer most. The primary functions they use are text production and information retrieval. They mainly learn the relevant knowledge and skills through self-media and knowledge-sharing platforms. Among the four typical task scenarios, GenAI is widely used in course learning and research activities, while its application in daily life and job search is relatively limited. The analysis of demographic variables shows that grade and major have a significant impact on university students’ use of GenAI. In addition, university students suggest that universities should offer relevant courses or lectures and provide comprehensive technical support to improve the popularity and operability of GenAI. This study provides suggestions for universities, education administration departments, and technology development departments to improve GenAI services. It will help universities optimize the allocation of educational resources and promote educational equity for sustainability. Full article

Due to their efficiency, stability, and enhanced language comprehension and analysis capabilities, generative AIs have attracted increasing attention in the field of writing as higher-level automated writing evaluation (AWE) feedback tools. However, few studies have examined the impact of pre-service teachers using generative AI in combination with their own teaching experience to provide feedback on Chinese writing. To fill this gap, based on 1035 writing feedback texts, we examined the differences in writing feedback between 11 pre-service teachers and Erine Bot (a generative AI) and interviewed the pre-service teachers about their willingness to cooperate with generative AI. The collaborative writing feedback generated by the pre-service teachers using AI was compared with the feedback generated by the pre-service teachers and generative AI separately. We identified that, although Ernie Bot provided significantly better feedback than the pre-service teachers in three specific areas (except for language expression), and both Ernie Bot and the pre-service teachers had respective advantages in terms of writing strategy, human–computer cooperative writing feedback was significantly better than the writing feedback provided by either Ernie Bot or the pre-service teachers alone. The was true across all aspects of the feedback in terms of focus and strategy. These findings can support the training of pre-service teachers and improve the writing quality of their students via implementing AI to provide more effective writing feedback. Full article

Background/Objectives: Background: Early-onset Alzheimer’s disease (EOAD) is primarily inherited in an autosomal dominant pattern, with mutations in the APP, PSEN1, and PSEN2 genes being central contributors. Diagnosing Alzheimer’s poses challenges due to the coexistence of various co-pathologies, and treatment options remain limited for most patients, apart from familial cases linked to specific genetic mutations. While significant research on Alzheimer’s genetics has been conducted in both Asian and Caucasian populations, the specific mutations and their clinical impacts in EOAD are still inadequately explored. This review aims to provide a detailed analysis of commonly reported genetic mutations and associated clinical features in EOAD patients from Asian and Western populations. Methods: Following the PRISMA-ScR guidelines, a systematic database search was conducted for studies published between 2016 and 2023. After screening 491 records, 36 studies from Asian cohorts and 40 from Western cohorts met the inclusion criteria. Results: The analysis revealed 127 unique mutations in the Asian population and 190 in the Western population. About 16.7% of Asian and 21.9% of Western studies covered both familial and sporadic AD, with consistent patterns across groups. Some mutations were shared between the populations and displayed similar clinical features, while others were population-specific. Conclusions: These findings underscore the considerable variability in EOAD mutations and phenotypes, emphasizing the importance of genetic testing in younger patients to enhance diagnostic accuracy and guide treatment strategies effectively. Full article

The extraction of disaster geospatial intelligence (DGI) from social media data with spatiotemporal attributes plays a crucial role in real-time disaster monitoring and emergency decision-making. However, conventional machine learning approaches struggle with semantic complexity and limited Chinese disaster corpus. Recent advancements in large language models (LLMs) offer new opportunities to overcome these challenges due to their enhanced semantic comprehension and multi-task learning capabilities. This study investigates the potential application of LLMs in disaster intelligence extraction and proposes an efficient, scalable method for multi-hazard DGI extraction. Building upon a unified ontological framework encompassing core natural disaster elements, this method employs parameter-efficient low-rank adaptation (LoRA) fine-tuning to optimize open-source Chinese LLMs using a meticulously curated instruction-tuning dataset. It achieves simultaneous identification of multi-hazard intelligence cues and extraction of disaster spatial entity attributes from unstructured Chinese social media texts through unified semantic parsing and structured knowledge mapping. Compared to pre-trained models such as BERT and ERNIE, the proposed method was shown to achieve state-of-the-art evaluation results, with the highest recognition accuracy (F1-score: 0.9714) and the best performance in structured information generation (BLEU-4 score: 92.9649). Furthermore, we developed and released DGI-Corpus, a Chinese instruction-tuning dataset covering various disaster types, to support the research and application of LLMs in this field. Lastly, the proposed method was applied to analyze the spatiotemporal evolution patterns of the Zhengzhou “7.20” flood disaster. This study enhances the efficiency of natural disaster monitoring and emergency management, offering technical support for disaster response and mitigation decision-making. Full article