最新刊期
- “在云数据中心资源消耗预测领域，专家提出了CNN-TCAM-BiGRU预测方法，有效提高了预测准确性，为资源管理和节能优化提供技术方案。”摘要：Cloud data center resource consumption prediction involves predicting energy consumption, computing resource requirements, and other factors of the data center. In order to improve the accuracy of resource consumption prediction in cloud data centers, so as to better manage resources and optimize energy saving effect, a novel prediction method CNN-TCCAM-BIGRU is proposed, which integrates multi-channel convolutional neural network with bidirectional gated cycle unit and time-channel attention mechanism. The method first uses signal decomposition technology to preprocess the original data, aiming at eliminating noise and highlighting key features in the data. Subsequently, two focused modules are constructed along the channel and time dimensions, and then an intermediate feature map is then built using these two modules.This intermediate feature map focuses on capturing key frequency bands and semantically relevant time periods. Compared with existing benchmark algorithms, it is verified that the proposed method shows remarkable performance advantages in three main evaluation indexes. CNN-TCAM-BiGRU effectively enhances the accuracy of resource consumption prediction in cloud data centers, offering an effective technical plan for resource management and energy-saving optimization in cloud data centers.关键词：temporal attention mechanism;variational mode decomposition;bidirectional gated recurrent unit;cloud data center51|13|0更新时间：2025-12-05
- “最新研究揭示没食子酸甲酯消除羟基自由基的机制，可作为有效的羟自由基清除剂。”摘要：The reaction mechanism of the scavenging of hydroxyl radical(·OH) by methyl gallate(MG) in physiological environment (aqueous phase at 310.15 K and 1.013×105 Pa) was investigated by using the density functional theory methods M06-2X and MN15 in combination with the SMD model of the self-consistent reaction field theory at the two level of SMD/MN15/6-311++G(4df,3pd)//SMD/M06-2X/6-311+G(d,p). There are three reaction channels for MG scavenging ·OH: H extracting by ·OH, addition of ·OH to unsaturated C, and single electron transfer from MG to ·OH. The calculations show that in the extraction channel, the reaction of ·OH extracting hydroxyl H is the most advantageous, which is a barrier-free and significantly exothermic process; the reaction of ·OH extracting methyl H is a subdominant reaction, which is an exothermic process with a free energy barrier of 37.8 to 41.2 kJ/mol. In the addition reaction channel, the addition of ·OH with unsaturated C is an exothermic process with a free energy barrier of 0.3 to 47.3 kJ/mol; the single-electron transfer from MG to ·OH is a slightly endothermic process with a free energy barrier of 42.1 kJ/mol. The results indicate that MG can eliminate ·OH radicals through three pathways: extraction H, addition, and electron transfer, and MG can be a good scavenger for ·OH radicals.关键词：methyl gallate;hydroxyl radical;density functional theory;self-consistent reaction field theory;transition state;electron transfer;free energy barrier20|3|0更新时间：2025-12-02
- “在自动驾驶领域，研究者提出了基于区域注意力的轻量级多任务全景驾驶感知网络APDP-Net，显著降低了计算开销，保持了出色的预测性能，为端侧自动驾驶感知提供了切实可行的解决方案。”摘要：Panoptic driving perception aims to simultaneously accomplish three perception tasks—traffic object detection, drivable area segmentation, and lane line detection—to meet the requirements of drivability prediction in autonomous driving. Existing methods usually adopt multi-task learning networks to process multiple tasks simultaneously, which have achieved certain progress in performance but generally suffer from high computational complexity as well as an imbalance between prediction accuracy and real-time performance. To address these problems, a lightweight multi-task panoramic driving perception network based on area attention, named APDP-Net, is proposed. First, the Residual-Efficient Layer Aggregation Network (R-ELAN) is introduced as the backbone network, in which the area attention module, characterized by a local and continuous attention mechanism, maintains the receptive field while reducing computational cost. Second, a GhostFPN-ELAN neck network with a feature pyramid structure is designed, introducing Ghost Convolution to reduce computation by reasonably compressing the feature space. Finally, for the lane line detection task, a Polarized Self-Attention mechanism is introduced to establish long-range connections between pixel-level features, enhancing the spatial feature representation capability of the network. Experimental results show that APDP-Net, through lightweight network design, significantly reduces computational overhead while maintaining excellent prediction performance, providing a feasible solution for edge-side autonomous driving perception.关键词：multi-task learning;panoptic driving perception;area attention;autonomous driving31|19|0更新时间：2025-12-01
- “在人工智能领域，专家建立了深度学习体系，为智能技术发展提供新方向。”
- “在社会治理领域，专家构建了社区治理情景知识图谱，显著提升了治理效率和智能化水平。”摘要：Knowledge graphs, due to their openness, flexibility, and strong interoperability, have significant application value in solving various social governance problems. Currently, community governance faces frequent resident demands and low processing efficiency, primarily due to the lack of systematic analysis and effective cross-departmental coordination mechanisms. To address this issue, this paper proposes a new method for constructing contextual knowledge graphs, aimed at improving community governance efficiency. First, by fine-tuning a pre-trained model, key governance elements are efficiently extracted from a small amount of labeled data, significantly improving data processing efficiency. Secondly, by combining risk-cause chain analysis with spatial-temporal dimensions, a new community governance ontology (CGO) model is constructed, which provides a structured semantic framework for the contextual knowledge graph. Based on this, the constructed contextual knowledge graph (CKG) integrates spatial-temporal information and inference rules, enabling precise risk-cause analysis and emergency severity prediction, while supporting intelligent decision-making inference and visualization queries. Experimental results show that the constructed contextual knowledge graph has strong spatial-temporal interoperability and inference capabilities, significantly enhancing the intelligence level and governance efficiency of community governance.关键词：risk causal chain;model fine-tuning;spatiotemporal interoperability;inference rules;contextual knowledge graph33|10|0更新时间：2025-11-10
- “遥感技术在极端干旱区植被生物量估算领域取得新进展，专家利用Sentinel-2影像和机器学习算法，为干旱区植被生物量估测提供新方法。”摘要：Using remote sensing technology to estimate the aboveground biomass of vegetation in extremely arid areas is of great significance for regional desertification monitoring and sustainable development. In this study, the core area of oasis in Turpan City was taken as the research area. Based on Sentinel-2 image and field sample data, vegetation index and band variables were extracted, and three sets of variable combinations were constructed. The traditional statistical model and four machine learning algorithm models of random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost) and K nearest neighbor (KNN) were compared to realize the estimation and spatial inversion of vegetation aboveground biomass in the core area of oasis in Turpan City. The results showed that in the traditional statistical model, the cubic curve equation established by the modified soil adjusted vegetation index (MSAVI) had the best fitting effect. The estimation accuracy of the machine learning model is better than that of the traditional statistical model as a whole. In the vegetation aboveground biomass estimation model constructed by the three variable combinations, the full variable combination composed of vegetation index and band variables has achieved the best estimation effect. The RF model has the highest estimation accuracy, with a coefficient of determination (R2) of 0.72, an average absolute error (MAE) of 105.64 g·m-2, and a root mean square error (RMSE) of 127.73 g·m-2. The total value of vegetation aboveground biomass in the oasis core area of Turpan City is about 1.599×106 kg, which has obvious spatial heterogeneity and shows a “fan” shape distribution. It shows that the oasis is larger than the desert, and the oasis gradually decreases to the hinterland of the desert. The RF model combined with Sentinel-2 image showed high accuracy and stability in the inversion of vegetation aboveground biomass in extremely arid areas, and could be used for the estimation of vegetation aboveground biomass in arid areas关键词：aboveground biomass of vegetation;machine learning;extremely arid areas;Sentinel-223|13|0更新时间：2025-11-10
- “在跨链身份互认领域，研究者设计了CAMRF机制，通过改进PageRank算法和BLS签名技术，有效提高了跨域身份认证的安全性和可靠性。”摘要：To address the challenge of identity authentication between entities across different administrative domains, a cross-blockchain authenticating mechanism based on random ferrying (CAMRF) was proposed in the present study. The mechanism first selects high-reputation candidate ferry nodes from ordinary nodes through the improved PageRank algorithm. Then, the consensus random number is generated based on the verifiable random function, and the ferry node group is dynamically elected as the notary group. Finally, the group nodes forward, sign and authenticate cross-domain messages, and use BLS(Boneh-Lynn-Shacham) lightweight aggregate signature technology to verify the authenticity and validity of the messages, while addressing issues related to centralization and security. Both theoretical analysis and experimental results indicate that CAMRF markedly enhances interoperability and strengthens the security and reliability of cross-domain authentication, thereby providing some theoretical significance and high practical value.关键词：cross-blockchain technology;identity authentication;verifiable random function(VRF);BLS signature;aggregate signatures28|11|0更新时间：2025-11-05
- 摘要：This study investigates the I-optimal design problem for two-response Scheffé canonical first-order and second-order mixture models. We derive the minimum trace of the average prediction variance matrix and provide the I-optimal measures for models with specific numbers of components. The results demonstrate that the I-optimal design for these two-response Scheffé mixture models is independent of the response covariance matrix.关键词：mixture experiments;multi-response;Scheffé model;I-optimal design43|10|0更新时间：2025-10-30
- 摘要：Taking the conjugative transfer of antibiotic resistance genes (ARGs) as an example for horizontal gene transfer, the effects of zinc oxide nanoparticles (ZnONPs) and oxytetracycline (OTC) co-exposure on the ARGs conjugative transfer frequency within the same bacterial genus (intra-genus) and across different genera (inter-genus) were investigated. Further, the mechanisms for co-exposure influences the conjugative transfer were explored. The results showed that the co-exposure of 1 mg/L ZnONPs and 20 μg/L OTC increased intra-genus and inter-genus conjugation frequencies by 2.08-fold and 2.34-fold, respectively, which was significantly higher than the individual exposures. The co-exposure elevated intracellular reactive oxygen species (ROS) generation, enhanced cell membrane permeability, promoted extracellular polymeric substance secretion, and upregulated conjugation-related genes while downregulating negative regulatory genes. These results indicate that the nanomaterials and antibiotics co-residue in environments may exacerbate the risk of bacterial resistance transmission.关键词：antibiotic resistance gene;conjugative transfer;nanomaterials;zinc oxide;oxytetracycline;co-exposure;antibiotic resistance21|10|0更新时间：2025-10-29
- “最新研究提出一种结合物理模型和深度学习的锂电池荷电状态估计方法，有效提升模型在复杂工况下的建模能力，实验结果表明其预测精度和鲁棒性优于现有模型。”摘要：Accurate estimation of the State of Charge (SOC) is fundamental to ensuring the reliable operation of lithium-ion batteries. To address the issue of insufficient input features in existing deep learning methods, this paper proposes an SOC estimation method based on a combination of a physical model and a deep learning algorithm. This method exploits the local feature extraction capability of the Convolutional Neural Network (CNN) and the temporal sequence processing ability of the Bidirectional Gated Recurrent Unit (BiGRU). By introducing the terminal voltage output from a first-order RC model as an input feature, which is combined with the measured voltage and current to form the neural network input, the modeling capability of the CNN-BiGRU under complex dynamic operating conditions is enhanced. Experimental results demonstrate the good SOC estimation performance of the CNN-BiGRU model. For the University of Maryland dataset, the Root Mean Square Error is 0.16% and the Mean Absolute Error is 0.12% at room temperature (25 ℃), outperforming existing models. Furthermore, the proposed model exhibits high prediction accuracy and robustness for lithium-ion batteries under varying ambient temperatures and different degradation levels.关键词：lithium-ion battery;SOC estimation;physical model;CNN-BiGRU model29|15|0更新时间：2025-10-28
- “在电动汽车续航性能和安全领域，专家建立了融合LSTM和Informer的锂离子电池SOH估计方法，为电池健康状态评估提供解决方案。”摘要：Lithium-ion batteries are prone to capacity degradation during long-term operation, which significantly affects the driving range and safety of electric vehicles. To accurately estimate the battery’s state of health (SOH), this study proposes a hybrid SOH estimation method that integrates a Long Short-Term Memory (LSTM) network with an Informer architecture. The Local Outlier Factor (LOF) algorithm is employed to preprocess the charging data of experimental batteries, and the maximum tangent angle of the charging voltage curve and its corresponding time are extracted as health features. The correlation between these health features and SOH is analyzed using the Spearman correlation coefficient. By combining the advantage of LSTM in capturing local temporal dependencies with the Informer’s ability to model global relationships, an LSTM–Informer serial network is constructed. The extracted health features are used as model inputs to achieve accurate SOH estimation. Experimental results demonstrate that the proposed method achieves high estimation accuracy, with the maximum absolute error maintained within 2.5%, and both the root mean square error and mean absolute error within 1%. Compared with traditional single-networks, the proposed method exhibits superior estimation performance and generalization capability.关键词：lithium-ion battery;state of health (SOH);outlier handling;Long Short-Term Memory (LSTM) networks;Informer network30|15|0更新时间：2025-10-27
- “在固体氧化物燃料电池系统故障识别领域，专家提出了融合聚类和RBF-SVM的故障识别方法，有效提升了故障识别能力，为解决系统故障识别问题提供解决方案。”摘要：To accurately identify faults in Solid Oxide Fuel Cell (SOFC) systems and overcome the challenges of traditional methods such as difficulty in parameter determination, high result randomness, and limitations in complex data processing, a fault identification method based on fusion clustering and Radial Basis Function Support Vector Machine (RBF-SVM) is proposed. This method employs a clustering algorithm based on fusion mechanism to process the performance degradation fault data of SOFC system reformer catalyst, obtaining clustering labels under different fault levels, and then uses the RBF-SVM algorithm to train the classified fault data to develop a fault identification model. This model can map fault feature information to higher-dimensional spaces to enhance fault identification capabilities and is further applied to identify air and fuel leakage faults. Experimental results demonstrate that the proposed fusion clustering algorithm effectively classifies faults, and the RBF-SVM fault identification model achieves 99.3% accuracy in identifying reformer catalyst degradation faults, 99.9% accuracy for air leakage faults, and 99.4% accuracy for fuel leakage faults.关键词：solid oxide fuel cell (SOFC);faults identification;fusion clustering;radial basis function support vector machine (RBF-SVM)37|11|0更新时间：2025-10-21
- “在新型储能技术领域，我国研究正由技术突破迈向系统协同与智能化发展，为新型电力系统构建提供关键支撑。”摘要：To reveal the research landscape and development trajectories of typical emerging energy storage technologies in China, this study focuses on five representative types: battery energy storage, molten salt energy storage, compressed air energy storage, flywheel energy storage, and hydrogen energy storage. A comparative bibliometric analysis was conducted based on core literature from the CNKI and Web of Science (WOS) databases during 2015-2025. The results indicate that, overall, domestic research is characterized by engineering practice and policy orientation, whereas international studies emphasize material mechanisms and theoretical innovation, reflecting a differentiated evolution of research pathways. Among them, battery energy storage exhibits the highest research intensity: domestic studies focus on state-of-charge (SOC) estimation, safety management, and system operation, while international research concentrates on solid electrolytes and post-lithium material systems. Molten salt energy storage has evolved from solar-thermal coupling toward multi-temperature thermochemical applications, emphasizing heat transfer efficiency and thermal performance improvement. Compressed air energy storage research centers on thermo-hydro-mechanical modeling and heat recovery optimization, showing strong potential for long-duration peak regulation. Flywheel energy storage has shifted from structural and control optimization to system integration and inertia support, and hydrogen energy storage demonstrates multi-energy coupling among wind, solar, and hydrogen systems, highlighting progress in hydrogen production efficiency. Overall, China’s research on emerging energy storage technologies is transitioning from single-technology breakthroughs toward system-level synergy and intelligent development. The complementary relationship between short-term power-type and long-duration capacity-type storage technologies is becoming clearer, with future progress expected to move toward higher safety, lower cost, longer lifespan, and multi-energy integration-providing critical support for the construction of a new-type power system.关键词：emerging energy storage;battery energy storage;molten salt energy storage;compressed air energy storage;flywheel energy storage;hydrogen energy storage;bibliometric analysis112|30|0更新时间：2025-10-17
- “在电动汽车充电领域，专家提出了一种新策略，优化充电成本和速度，减少经济损失，提升充电效率。”摘要：Most of the existing charging methods primarily focus on charging speed and charging safety, ignoring the economic loss of the user during the charging process. To address this limitation, this paper proposes a fast-charging strategy that takes into account battery degradation and time-of-use electricity price. The time-of-use electricity price serves as an input parameter, the battery degradation and power loss are converted into charging economic loss, which is optimized by using the reinforcement learning algorithm and the electrochemical-thermal-aging coupling model. The results show that the proposed method can autonomously adjust the charging rate according to the time-of-use electricity price. Given the same charging speed, this method reduces economic loss by 20.9% compared with the constant current constant voltage charging method and 15.3% compared with the multi-stage constant current charging method. Under the same physical constraints, this method is 21.1% faster than the pulse charging method and 9.9% faster than the constant current constant voltage method, optimizing both charging cost and charging speed.关键词：lithium-ion battery;time-of-use electricity price;battery degradation;fast charging strategy48|23|0更新时间：2025-10-17
- “在乡村振兴和健康中国政策背景下，成都和上海8个农村新型社区的研究表明，物质环境设施、交通条件、社会生活环境满意度显著正向影响居民自评健康，为相关政策制定和资源配置提供参考。”摘要：Under the background of rural revitalization and healthy China policy, the research on the subjective perception of rural environment and residents’ self-rated health is increasing. Existing studies mostly focus on single indicators, and lack of discussion on the joint effect of group indicators. Taking 8 typical new rural communities in Chengdu and Shanghai as examples, this study conducted a questionnaire survey on individual villagers and obtained 493 valid sample data. Factor analysis method and stepwise regression model were used to explore the common factors affecting residents’ subjective environmental perception in new rural communities and their mechanism. The results showed that the comprehensive common factor of satisfaction with physical and environmental facilities and health management (F1), the comprehensive common factor of satisfaction with traffic conditions and interpersonal activities (F2), and the comprehensive common factor of satisfaction with social life environment (F3) all significantly positively affected the self-rated health of residents, and the F1 had the greatest influence. Based on the three comprehensive common factors, the promotion strategy of planning and design of new rural communities for health was proposed, which could provide reference for relevant policy making and resource allocation of new rural communities.关键词：subjective perception;self-rated health;factor analysis method;design strategy30|38|0更新时间：2025-10-11
- “在因果推断领域，专家提出了一种三阶段加权因果效应估计算法，有效解决了协变量缺失问题，为因果效应估计提供了新方案。”摘要：Faced with the challenges of obtaining randomized controlled trial data, causal inference based on observational data emerges as an alternative research paradigm. However, missing covariates in observational data often introduces bias, leading to distorted causal effect estimation. To solve this problem, this paper proposed a three-stage weighted causal effect estimation algorithm with missing covariates. First, covariates were categorized into confounders, exposure predictors and outcome predictors based on the causal graph, with multiple imputation implemented for missing confounders to preserve the integrity of causal structures. Second, a missing pattern was introduced, constructing propensity score models through an improved covariate balancing strategy. Finally, causal effects were estimated by integrating inverse probability weighting across missing patterns. Simulation results demonstrate that compared with gradient boosting machines, the proposed method reduces root mean square error in most scenarios, validating its effectiveness in estimating causal effects under missing data conditions.关键词：causal inference;missing data for covariates;treatment effect;propensity score76|25|0更新时间：2025-10-09
- “据最新研究，三峡库区香溪河VOCs分布特征及环境健康风险评估显示，水体和沉积物未对大气VOCs污染产生贡献，土壤中农药残留1，2，4-三氯苯对大气VOCs污染产生贡献。”摘要：In order to reveal the distribution and characteristics of volatile organic compounds (VOCs) along typical tributaries of the Three Gorges Reservoir Region, evaluate the environmental impact and health risks of VOCs, 8 sites were set up along the Xiangxi River in 2021 (March, June, September, and December), the samples of air, water, soil (altitude 175-190 m), and sediment(altitude below 145 m) were collected, 59 VOCs were determined, the ·OH loss rate(R) and ozone formation potential(OFP) were calculated, and the hazard index (HI) and lifetime carcinogenic risk (LCR) of VOCs were evaluated. The results showed that VOCs was not detected in all samples of water and sediment, VOCs of most soil samples was not detected in summer and autumn, 1,2,3-trichlorobenzene (0.35-0.41 μg/kg) and 1,2,4-trichlorobenzene(0.39-0.52 μg/kg) of all soil samples were detected in spring, 1,2,4-trichlorobenzene(0.34-0.42 μg/kg) was detected in winter, the spatial distribution of 1,2,3-trichlorobenzene and 1,2,4-trichlorobenzene were upstream, downstream, tributaries, and midstream from large to small. The average concentration of atmospheric VOCs was 14.69±9.1 μg/m3, with aromatic hydrocarbons (56.3%), oxygenated volatile organic compounds (25.2%), halogenated hydrocarbons (18.5%). from large to small, the seasonal variation was spring, autumn, summer,and winter, the spatial variation were upstream, midstream, downstream, tributaries. most of the top 7 substances about the contribution rankings of R and OFP were aromatic hydrocarbons at different sites and four seasons. The HI was less than 1 at every site, and the range of LCR was from 6.1×10-6 to 9.6×10-6. The research has shown that there was not contribution to atmospheric VOCs pollution in water and sediments, while pesticide residues such as 1,2,4-trichlorobenzene of soil contributed to atmospheric VOCs pollution, aromatic hydrocarbons were the main active components for contribution species of R and OFP. The HI were all at an acceptable safety level, the LCR were all in grade Ⅱ of low probability. The study can provide theoretical reference for controlling VOCs emissions and reducing their environmental impact and health risks along the Xiangxi River.关键词：Xiangxi River;volatile organic compounds (VOCs);·OH loss rate;ozone formation potential;health risk22|19|0更新时间：2025-10-09
- “在新能源传输领域，专家采用CFD软件分析海底电缆电磁场分布，为保护海洋生态提供新思路。”摘要：Submarine cables served as critical new energy transmission infrastructure connecting land and sea, and the electromagnetic radiation generated during their operation exerted significant impacts on the marine ecological environment. To address this issue, this study employs Computational Fluid Dynamics (CFD) software and finite element simulation to conduct an in-depth analysis of the electromagnetic field distribution patterns in high voltage alternating current submarine cables. Simulation accuracy is enhanced through refined mesh division and optimization of material parameters, while a 2D axisymmetric model is constructed to simplify calculations. The results demonstrate that: the surface potential of the cable exhibits significant spatial heterogeneity, with the electric field intensity reaching 1.27×10⁵ V/m on the inner side of the insulation layer and a local minimum of -6.33×10⁴ V/m in the core region, reflecting the localized characteristics of strong electric field sources; the magnetic field is distributed in concentric circles centered on the cable core, with the central magnetic flux density significantly higher than that in the peripheral areas, showing a gradient decay with increasing distance; the ferromagnetic properties of the armor layer cause magnetic field distortion, and the phase differences of a three-phase current further exacerbate the spatial non-uniformity of the magnetic field.关键词：finite element simulation;2D axisymmetric model;three-core submarine cable;electromagnetic radiation;electromagnetic field characteristics74|24|0更新时间：2025-10-09
- “在电池管理领域，研究人员提出了基于CL-NARX神经网络的SOC估计模型，有效提高了估计精度和鲁棒性，为锂离子电池安全运行提供技术支持。”摘要：To address the problems of insufficient estimation accuracy, error accumulation, and poor robustness of traditional State of Charge (SOC) estimation models under actual operating conditions, this paper proposes a Closed-loop Nonlinear Autoregressive eXogenous (CL-NARX) neural network model to improve SOC estimation accuracy. The model enhances the fitting capability for complex battery processes by introducing a closed-loop feedback mechanism, effectively suppressing error accumulation, and strengthening robustness by optimizing key hyperparameters. The experimental results show that the model achieves optimal performance, when the training iteration number is 150, the number of hidden layers is 10, the input delay layers is 5, and the output delay layers is 2, with estimation errors significantly superior to other neural network models. The maximum error, RMSE, MAE, and MAPE are reduced to 2.58%, 1.41%, 1.36%, and 4.57%, respectively. The model demonstrates high accuracy, effective error handling, and strong robustness, providing reliable technical support for the safe operation of lithium-ion batteries.关键词：lithium-ion battery;State of Charge (SOC);Closed-Loop Nonlinear Autoregressive eXogenous (CL-NARX);backpropagation (BP) neural network41|28|0更新时间：2025-10-09
- “在情感分析领域，专家构建金融情感知识图谱并结合少样本提示对大语言模型进行多层级知识增强，显著提升金融文本情感分析准确性，为金融领域情感分析提供新思路。”摘要：In the field of sentiment analysis, large language model (LLM) have been widely adopted for sentiment judgment. However, financial texts inherently involve complex sentiment features with blurred boundaries between distinct sentiment categories, presenting LLM with substantial challenges in accurate classification. To address this issue, we proposed a method that integrates a financial sentiment knowledge graph with few-shot prompting for multi-level knowledge enhancement. The graph was semi-automatically constructed using LLM, and financial terms within it were annotated with sentiment labels. This information was explicitly represented through prompt templates, providing semantic-level support for the LLM. Experiments were conducted on Qwen2.5-7B, ChatGLM3-6B, and ChatGLM4-9B, demonstrating that our approach significantly improves sentiment analysis accuracy, with an average F1 score increase of 3.67 percentage points compared to LLM that don’t employ this method. The results show that this method provides new solutions and important references for financial text sentiment analysis.关键词：financial text sentiment analysis;knowledge-enhanced;LLM(large language model);knowledge graph;few-shot prompting89|40|0更新时间：2025-09-18
- Tel：027-68756952 Email：whdz@whu.edu.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
- Total Visits：1300706 Visits Today：611
Bulk Citation
0