{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T12:26:23Z","timestamp":1787574383837,"version":"build-2736575974"},"reference-count":54,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2026,8,22]],"date-time":"2026-08-22T00:00:00Z","timestamp":1787356800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>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\u2013time 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 R2 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.<\/jats:p>","DOI":"10.3390\/s26175322","type":"journal-article","created":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T10:07:53Z","timestamp":1787566073000},"page":"5322","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Reliable Hardware Sensor and Large Language Model Fusion for Intelligent Short-Term Market Risk Sensing and Prediction"],"prefix":"10.3390","volume":"26","author":[{"given":"Zijian","family":"Zhou","sequence":"first","affiliation":[{"name":"Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nuo","family":"Wang","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"},{"name":"Capital University of Economics and Business, Beijing 100070, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shengzhe","family":"Xu","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Surui","family":"Hua","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hanyang","family":"Wang","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"},{"name":"China Agricultural University, Beijing 100083, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yachi","family":"Liu","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"},{"name":"China Agricultural University, Beijing 100083, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Manzhou","family":"Li","sequence":"additional","affiliation":[{"name":"Peking University, Beijing 100871, China"},{"name":"China Agricultural University, Beijing 100083, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,22]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"161","DOI":"10.37256\/ccds.7120269128","article-title":"Intelligent Agent-Based Market Research: Cloud-Orchestrated Large Language Models as Financial Analysts","volume":"7","author":"Chen","year":"2026","journal-title":"Cloud Comput. Data Sci."},{"key":"ref_2","first-page":"4","article-title":"Adaptive AI-driven on-demand packet inspection for low-latency market data connectivity between peer exchanges","volume":"31","author":"Balakrishnan","year":"2023","journal-title":"J. Comput. Anal. Appl."},{"key":"ref_3","first-page":"31","article-title":"Under pressure: Market conditions and stress","volume":"19","author":"Aldasoro","year":"2022","journal-title":"BIS Q. Rev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"20010","DOI":"10.1109\/JSEN.2022.3203994","article-title":"An RFID-based wireless vibration and physical-shock sensing system using edge processing","volume":"22","author":"Song","year":"2022","journal-title":"IEEE Sens. J."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"21813","DOI":"10.1109\/JSEN.2024.3379222","article-title":"Spatial environment perception and sensing in automated systems: A review","volume":"24","author":"Fei","year":"2024","journal-title":"IEEE Sens. J."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1080\/09540091.2021.1940101","article-title":"S_I_LSTM: Stock price prediction based on multiple data sources and sentiment analysis","volume":"34","author":"Wu","year":"2022","journal-title":"Connect. Sci."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Qian, C., Zhang, H., Sha, L., and Zheng, Z. (2025). Hsf: Defending against jailbreak attacks with hidden state filtering. Companion Proceedings of the ACM on Web Conference 2025, Association for Computing Machinery.","DOI":"10.1145\/3701716.3717659"},{"key":"ref_8","first-page":"64","article-title":"Study of market indicators used for technical analysis","volume":"12","author":"Dongrey","year":"2022","journal-title":"Int. J. Eng. Manag. Res."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1159","DOI":"10.1109\/JPROC.2023.3309299","article-title":"Toward label-efficient emotion and sentiment analysis","volume":"111","author":"Zhao","year":"2023","journal-title":"Proc. IEEE"},{"key":"ref_10","first-page":"99","article-title":"Forecasting Hardware Failures or Resource Bottlenecks Before They Occur","volume":"3","author":"Karri","year":"2022","journal-title":"Int. J. Emerg. Res. Eng. Technol."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"e2111997119","DOI":"10.1073\/pnas.2111997119","article-title":"Toward data-driven, dynamical complex systems approaches to disaster resilience","volume":"119","author":"Yabe","year":"2022","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Abdelsamie, M., and Wang, H. (2024). Comparative analysis of LLM-based market prediction and human expertise with sentiment analysis and machine learning integration. Proceedings of the 2024 7th International Conference on Data Science and Information Technology (DSIT), IEEE.","DOI":"10.1109\/DSIT61374.2024.10881868"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2427389","DOI":"10.1155\/2022\/2427389","article-title":"Stock market prediction based on financial news text mining and investor sentiment recognition","volume":"2022","author":"Bi","year":"2022","journal-title":"Math. Probl. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Herrera-Poyatos, D., Pel\u00e1ez-Gonz\u00e1lez, C., Zuheros, C., Herrera-Poyatos, A., Tejedor, V., Herrera, F., and Montes, R. (2025). An overview of model uncertainty and variability in LLM-based sentiment analysis: Challenges, mitigation strategies, and the role of explainability. Front. Artif. Intell., 8.","DOI":"10.3389\/frai.2025.1609097"},{"key":"ref_15","first-page":"201","article-title":"Anomaly detection in IoT sensor data using machine learning techniques for predictive maintenance in smart grids","volume":"5","author":"Omol","year":"2024","journal-title":"Int. J. Sci. Technol. Manag."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Fu, C., Wang, X., Zhang, Y., Xiong, Z., Pan, J., and Yin, J. (2026). AI-Driven Sensing for Cross-Lingual Risk Prediction via Semantic Alignment and Multimodal Temporal Fusion. Appl. Sci., 16.","DOI":"10.3390\/app16083741"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Kong, Y., Hwang, Y., Kaiser, M., Vryonides, C., Oomen, R., and Zohren, S. (2025, January 15\u201318). Fusing Narrative Semantics for Financial Volatility Forecasting. Proceedings of the 6th ACM International Conference on AI in Finance, Singapore.","DOI":"10.1145\/3768292.3771256"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"11","DOI":"10.4236\/jcc.2024.121002","article-title":"Lexicon and deep learning-based approaches in sentiment analysis on short texts","volume":"12","author":"Islam","year":"2024","journal-title":"J. Comput. Commun."},{"key":"ref_19","first-page":"22","article-title":"Financial news sentiment analysis and market sentiment prediction based on large language models","volume":"2","author":"Qi","year":"2025","journal-title":"J. Comput. Signal Syst. Res."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1145\/3533018","article-title":"Read the news, not the books: Forecasting firms\u2019 long-term financial performance via deep text mining","volume":"14","author":"Zhai","year":"2023","journal-title":"ACM Trans. Manag. Inf. Syst."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Farimani, S.A., Jahan, M.V., and Milani Fard, A. (2022). From text representation to financial market prediction: A literature review. Information, 13.","DOI":"10.3390\/info13100466"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Wu, Y., Sun, M., Zheng, H., Hu, J., Liang, Y., and Lin, Z. (2024). Integrative analysis of financial market sentiment using cnn and gru for risk prediction and alert systems. Proceedings of the 2024 International Conference on Electronics and Devices, Computational Science (ICEDCS), IEEE.","DOI":"10.1109\/ICEDCS64328.2024.00078"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Kumar, R., Kumar, H., and Shalini, K. (2025). Cross-Domain Knowledge Transfer using LLMs and Domain-Specific Knowledge Graphs. Proceedings of the 2025 IEEE International Conference on Emerging Technologies and Applications (MPSec ICETA), IEEE.","DOI":"10.1109\/MPSecICETA64837.2025.11118450"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1007\/s44443-025-00330-w","article-title":"LLM-enhanced multi-causal event causality mining in financial texts","volume":"37","author":"Yan","year":"2025","journal-title":"J. King Saud Univ. Comput. Inf. Sci."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Yigit, Y., Ferrag, M.A., Ghanem, M.C., Sarker, I.H., Maglaras, L.A., Chrysoulas, C., Moradpoor, N., Tihanyi, N., and Janicke, H. (2025). Generative AI and LLMs for critical infrastructure protection: Evaluation benchmarks, agentic AI, challenges, and opportunities. Sensors, 25.","DOI":"10.3390\/s25061666"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"4055","DOI":"10.1109\/TWC.2025.3535714","article-title":"Latent diffusion model-enabled low-latency semantic communication in the presence of semantic ambiguities and wireless channel noises","volume":"24","author":"Pei","year":"2025","journal-title":"IEEE Trans. Wirel. Commun."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Bin Mofidul, R., Alam, M.M., Rahman, M.H., and Jang, Y.M. (2022). Real-time energy data acquisition, anomaly detection, and monitoring system: Implementation of a secured, robust, and integrated global IIoT infrastructure with edge and cloud AI. Sensors, 22.","DOI":"10.3390\/s22228980"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"81","DOI":"10.31181\/dmame7220241040","article-title":"The application of embedded hardware system and blockchain in rural financial management cloud platform","volume":"7","author":"Sun","year":"2024","journal-title":"Decis. Mak. Appl. Manag. Eng."},{"key":"ref_29","first-page":"246","article-title":"Internet of things based wireless sensor network: A review","volume":"27","author":"Nourildean","year":"2022","journal-title":"Indones. J. Electr. Eng. Comput. Sci."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1109\/MSP.2023.3258060","article-title":"Twenty-five years of sensor array and multichannel signal processing: A review of progress to date and potential research directions","volume":"40","author":"Liu","year":"2023","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_31","unstructured":"Chowdhary, M.A.M. (2025). Financial Network Infrastructure: Scalability, Security and Optimization. [Master\u2019s Thesis, Metropolia University of Applied Sciences]."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Lekidis, A., and Papageorgiou, E.I. (2023). Edge-based short-term energy demand prediction. Energies, 16.","DOI":"10.3390\/en16145435"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"102768","DOI":"10.1016\/j.ipm.2021.102768","article-title":"DATA-driven shock impact of COVID-19 on the market financial system","volume":"59","author":"Li","year":"2022","journal-title":"Inf. Process. Manag."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Farzulla, M. (2025). Infrastructure vs. Regulatory Shocks: Asymmetric Volatility Response in Cryptocurrency Markets. Res. Sq.","DOI":"10.21203\/rs.3.rs-8323026\/v1"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Sk\u00e5lvik, A.M., Saetre, C., Fr\u00f8ysa, K.E., Bj\u00f8rk, R.N., and Tengberg, A. (2023). Challenges, limitations, and measurement strategies to ensure data quality in deep-sea sensors. Front. Mar. Sci., 10.","DOI":"10.3389\/fmars.2023.1152236"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"8303555","DOI":"10.1155\/2022\/8303555","article-title":"Financial asset risk measurement based on smart sensor big data security analysis and bayesian posterior probability model","volume":"2022","author":"Lu","year":"2022","journal-title":"Wirel. Commun. Mob. Comput."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"3563","DOI":"10.1109\/TCCN.2025.3612760","article-title":"Toward edge general intelligence with multiple-large language model (Multi-LLM): Architecture, trust, and orchestration","volume":"11","author":"Luo","year":"2025","journal-title":"IEEE Trans. Cogn. Commun. Netw."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Wang, M., Zhang, Q., Liu, X., Zhang, J., Yu, F., Zhang, X., and Zhao, R. (2025). Research progress on multimodal data fusion in forest resource monitoring. Front. Plant Sci., 16.","DOI":"10.3389\/fpls.2025.1710618"},{"key":"ref_39","first-page":"89","article-title":"Research on Sentiment Analysis Based on Multi-source Data Fusion and Pre-trained Model Optimization in Quantitative Finance","volume":"6","author":"Sun","year":"2025","journal-title":"SOcio-Econ. Stat. Res."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Cao, Y., Chen, Z., Kumar, P., Pei, Q., Yu, Y., Li, H., Dimino, F., Ausiello, L., Subbalakshmi, K., and Ndiaye, P.M. (2025). RiskLabs: Predicting financial risk using large language model based on multimodal and multi-sources data. Proceedings of the 2025 IEEE International Conference on Data Mining Workshops (ICDMW), IEEE.","DOI":"10.1109\/ICDMW69685.2025.00107"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"7918","DOI":"10.1287\/mnsc.2021.4261","article-title":"Demand estimation using managerial responses to automated price recommendations","volume":"68","author":"Garcia","year":"2022","journal-title":"Manag. Sci."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Lu, J. (2025). Time-Series Foundation Models in Finance: Pretraining Corpora, Architectures, Financial Benchmarks, and Risk-Aware Evaluation. Proceedings of the 2025 2nd International Conference on Digital Economy and Computer Science, Association for Computing Machinery.","DOI":"10.1145\/3785706.3785728"},{"key":"ref_43","unstructured":"Rao, H., Zhao, Y., Hou, X., Wang, S., and Wang, H. (2025). Software engineering for large language models: Research status, challenges and the road ahead. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"739","DOI":"10.1007\/s42765-024-00387-8","article-title":"Introducing rich heterojunction surfaces to enhance the high-frequency electromagnetic attenuation response of flexible fiber-based wearable absorbers","volume":"6","author":"Han","year":"2024","journal-title":"Adv. Fiber Mater."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Yang, H., Lambotharan, S., Derakhshani, M., and Hanzo, L. (2026). LLM-enhanced space-air-ground-sea integrated networks. IEEE Communications Magazine, IEEE.","DOI":"10.1109\/MCOM.001.2500470"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"215","DOI":"10.1111\/j.2517-6161.1958.tb00292.x","article-title":"The regression analysis of binary sequences","volume":"20","author":"Cox","year":"1958","journal-title":"J. R. Stat. Soc. Ser. B Stat. Methodol."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). Xgboost: A scalable tree boosting system. Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_48","first-page":"3146","article-title":"Lightgbm: A highly efficient gradient boosting decision tree","volume":"30","author":"Ke","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Graves, A. (2012). Long short-term memory. Supervised Sequence Labelling with Recurrent Neural Networks, Springer.","DOI":"10.1007\/978-3-642-24797-2"},{"key":"ref_50","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., and Zhang, W. (2021, January 2\u20139). Informer: Beyond efficient transformer for long sequence time-series forecasting. Proceedings of the AAAI Conference on Artificial Intelligence, Virtual.","DOI":"10.1609\/aaai.v35i12.17325"},{"key":"ref_52","unstructured":"Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2019, January 2\u20137). Bert: Pre-training of deep bidirectional transformers for language understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human\nLanguage Technologies, Volume 1 (Long and Short Papers), Minneapolis, MA, USA."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Shen, Y., and Zhang, P.K. (2024). Financial sentiment analysis on news and reports using large language models and finbert. Proceedings of the 2024 IEEE 6th International Conference on Power, Intelligent Computing and Systems (ICPICS), IEEE.","DOI":"10.1109\/ICPICS62053.2024.10796670"},{"key":"ref_54","unstructured":"Wu, S., Irsoy, O., Lu, S., Dabravolski, V., Dredze, M., Gehrmann, S., Kambadur, P., Rosenberg, D., and Mann, G. (2023). Bloomberggpt: A large language model for finance. arXiv."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/17\/5322\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T11:56:46Z","timestamp":1787572606000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/17\/5322"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,22]]},"references-count":54,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26175322"],"URL":"https:\/\/doi.org\/10.3390\/s26175322","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,22]]}}}