中国科学院数学与系统科学研究院期刊网
JSSI

28 April 2025, Volume 13 Issue 2
Previous      Next

Download cover Download table of contents
  • Select all
    |
  • AI and Smart Decision-Making: Theory, Modelling and Applications
    Hong ZHAO, Zongshui WANG, Stine Jessen HAAJONSSON
    Journal of Systems Science and Information. 2025, 13(2): 157-158.
    Abstract ( 76 ) Download PDF ( 57 )   Knowledge map   Save
  • Mapping the AI and Smart Decision-Making Research Landscape: A Bibliometric Analysis
    Danyang HE, Zongshui WANG
    Journal of Systems Science and Information. 2025, 13(2): 159-186. https://doi.org/10.12012/JSSI-2023-0030
    Abstract ( 70 ) Download PDF ( 38 )   Knowledge map   Save
    With the rapid development of artificial intelligence technology and its advantages in decision-making, smart decision-making is receiving increasing attention from both academics and practitioners. This paper uses a bibliometric approach to analyze 783 works of literature related to smart decision-making from 1965-2022 to understand the research content and development process in the field. Through co-citation analysis, this paper identifies critical publications and research clusters in the field. Combined with keyword analysis, this paper provides a systematic review of the current state of research on smart decision-making. On this basis, the conceptual framework of the field is presented. According to the existing literature and bibliometric analysis, this study provides a list of research questions from three dimensions that deserve further consideration, providing insights into the trend of using AI for decision-making in the future.
  • Consumer Privacy Risks and Privacy Protection in the Era of Artificial Intelligence
    Xia LIU, Hong ZHAO
    Journal of Systems Science and Information. 2025, 13(2): 187-202. https://doi.org/10.12012/JSSI-2023-0050
    Abstract ( 66 ) Download PDF ( 15 )   Knowledge map   Save
    Artificial intelligence has transformed marketing and consumer lives. However, despite its multiple beneficial effects, unprecedented artificial intelligence uses pose enormous challenges to consumer privacy protection. This research aims to provide a comprehensive picture of privacy research and enhance privacy protection. We first give more specific definitions of consumer privacy according to data analytic processes. Second, we investigate how AI-related factors, privacy risk, privacy protection, and other factors influence consumer privacy decision-making, and draw a conceptual model based on the extended Antecedents-Privacy Concerns-Outcomes (APCO) model. Third, we examine the potential risks of each data analytic step and the causes of privacy risks from consumer, company, and technology perspectives, and propose privacy protection schemes correspondingly. Lastly, we also provide agenda for future research.
  • CBSG_DPI — A Deep Learning Method for Drug-target Interactions Based on a Graph Convolutional Neural Network
    Yuyu GAN, Yanjia YU, Yongshang YU, Le ZHONG, Yong LIU
    Journal of Systems Science and Information. 2025, 13(2): 203-220. https://doi.org/10.12012/JSSI-2023-0122
    Abstract ( 74 ) Download PDF ( 18 )   Knowledge map   Save
    Drug-protein interaction (DPI) prediction in drug discovery and new drug design plays a key role, but the traditional in vitro experiments would incur significant temporal and financial costs, cannot smoothly advance drug-protein interaction research, so many computer prediction models have emerged, and the current commonly used is based on deep learning method. In this paper, a deep learning model Computer-based Drug-Protein Interaction CBSG_DPI is proposed to predict drug-protein interactions. This model uses the protein features extracted by the Computed Tomography CT and Bert method and the drug features extracted by the SMILES2Vec method and input into the graph convolutional neural network (GCN) to complete the prediction of drug-protein interactions. The obtained results show that the proposed model can not only predict drug-protein interactions more accurately but also train hundreds of times faster than the traditional deep learning model by abandoning the traditional grid search algorithm to find the best parameters.
  • Dissemination and Sentiment Analysis of Mega Sports Events Based on Big Data Statistics Under Social Media Environment
    Jiajia XIONG, Wei LIU, Jianwei MA
    Journal of Systems Science and Information. 2025, 13(2): 221-239. https://doi.org/10.12012/JSSI-2023-0132
    Abstract ( 71 ) Download PDF ( 9 )   Knowledge map   Save
    Given the rise of artificial intelligence, big data analytics has emerged as an important tool for processing and assimilating the enormous volume of data available on social media. It is of great theoretical and practical significance to explore the public opinion diffusion process and characteristics, and users' emotions of mega sports events based on big data statistics in the social media environment. This paper takes the Jakarta Asian Games, Russian World Cup and PyeongChang Winter Olympics held in 2018 as cases, uses text mining and social network analysis methods to analyze the dissemination process of social media users' data, presents the semantic words disseminated in sports events through high-frequency word cloud diagrams, and summarizes the general rules of public opinion dissemination. The results show that the more users' participation, the greater diffusion volume, and the diffusion process shows fast increasing, short duration, scattered topics, diversified contents, and strong guidance and weak continuity of attention. The high-frequency words, except for the names of the events, such as "cheer", "win the game" and "must win", have obvious concentration of emotional words.
  • Structural Transformation of the Financial Market of Ukraine in the Environment of Innovative Technologies of Metaspace
    Larysa VDOVENKO, Leonid MELNYK, Olena POLOVA, Olena MARTSENIUK, Oksana RUDA
    Journal of Systems Science and Information. 2025, 13(2): 240-273. https://doi.org/10.12012/JSSI-2024-0136
    Abstract ( 75 ) Download PDF ( 12 )   Knowledge map   Save
    This article investigates the structural transformation of Ukraine's financial market within the context of innovative metaspace technologies. It presents models outlining the structural changes in financial market sectors amid the rapid development of metaspace technologies, emphasizing the systematic prognostic nonlinear dependence of deposit, credit, and currency flows on time intervals within the FinTech services' payment platform. The article introduces system methods for modeling nonlinear processes and combinatorial procedures for organizing the environment of innovative metaspace technologies within the financial market. It systematically categorizes approaches for measuring the effectiveness signs of financial market structure development in the innovative metaspace. The article argues for the need to structure and evaluate the financial market sectors of Ukraine through non-combinatorial tasks, emphasizing the convenience of describing variables influencing market structure and visually depicting the metaspace environment as a simulated FinTech ecosystem. Standardizing performance indicator measurement algorithms for precise intervals of financial resource movement is advocated. The article establishes that combinatorial tasks for structuring financial market development objects in the metaspace environment should align with configurations influenced by variable factors, such as currency, deposit, and credit flows. An analysis of transactions involving currency exchange, deposits, and loans through FinTech services' payment platform in Ukraine's financial market is presented. Additionally, the article conducts a formalized assessment of changes in currency, deposit, and credit flow transactions influenced by the digital environment of FinTech services in Ukraine's financial market sectors.
  • Analysis and Constraint Cost Optimization for Discrete-time Queue with Bi-Level Randomized (p, N1, N2)-Policy
    Xinyu KUANG, Yinghui TANG, Shaojun LAN
    Journal of Systems Science and Information. 2025, 13(2): 274-298. https://doi.org/10.12012/JSSI-2024-0067
    Abstract ( 58 ) Download PDF ( 6 )   Knowledge map   Save
    This paper proposes a new discrete-time Geo/G/1 queueing model under the control of bi-level randomized (p, N1, N2)-policy. That is, the server is closed down immediately when the system is empty. If N1 (≥1) customers are accumulated in the queue, the server is activated for service with probability p (0≤ p≤1) or still left off with probability (1-p). When the number of customers in the system becomes N2 (≥ N1), the server begins serving the waiting customers until the system becomes empty again. For the model, firstly, we obtain the transient solution of the queue size distribution and the explicit recursive formulas of the stationary queue length distribution by employing the total probability decomposition technique. Then, the expressions of its probability generating function of the steady-state queue size and the expected steady-state queue size are presented. Additionally, numerical examples are conducted to discuss the effect of the system parameters on some performance indices. Furthermore, the steady-state distribution of queue length at epochs n-, n and outside observer's observation epoch are explored, respectively. Finally, we establish a cost function to investigate the cost optimization problem under the constraint of the average waiting time. And the presented model provides a less expected cost as compared to the traditional N-policy.
  • Digital Analysis of Stability and Safety in Heterogeneous Traffic Flow Mixed with Trucks and Connected and Autonomous Vehicles (CAVs)
    Weiqing ZHUANG, Yifan PEI
    Journal of Systems Science and Information. 2025, 13(2): 299-312. https://doi.org/10.12012/JSSI-2024-0056
    Abstract ( 57 ) Download PDF ( 6 )   Knowledge map   Save
    Connected and autonomous vehicles (CAVs) are expected to coexist alongside human-driven vehicles on roads for the foreseeable future. This study explores the stability and safety of mixed traffic streams, including traditional trucks and cars alongside CAVs. The study utilizes the intelligent driver model and cooperative adaptive cruise control model to characterize human-driven vehicles (including cars and trucks) and CAVs, respectively. It investigates how different ratios of trucks and penetration rates of CAVs impact the linear stability of mixed traffic flows and delineate their stability domains. Additionally, a simulation experiment is conducted using SUMO software to assess the safety implications of traffic congestion at on-ramp bottlenecks, specifically analyzing the safety dynamics of mixed traffic streams. The findings indicate that CAVs enhance both the stability and safety of mixed traffic flows. The presence of trucks is associated with reduced stability values at similar CAVs penetration rates. In scenarios without trucks, CAVs can elevate traffic safety by 58.28%-71.28%, whereas in the presence of trucks, although the enhancement diminishes, safety levels can still improve by 48.67%-65.11%.
  • Research on New Energy Vehicle Type Selection Method Based on Inversion
    Kai LAI, Songyuan DIAO, Yada HU, Quanyi LIU, Chunsheng CUI
    Journal of Systems Science and Information. 2025, 13(2): 313-324. https://doi.org/10.12012/JSSI-2024-0118
    Abstract ( 65 ) Download PDF ( 19 )   Knowledge map   Save
    This paper investigates the rank reversal issue in the selection of new energy vehicle types, focusing on consumers aged 20 to 30. It employs both the AHP and the PCbHA methods to rank four types of the new energy vehicles — pure electric vehicles, plug-in hybrid electric vehicles, range-extended electric vehicles, and fuel cell vehicles, based on ten influential factors: purchase cost, maintenance cost, fuel and electricity cost, safety, passability, endurance, appearance, brand power, power, and space. To verify the effectiveness of the PCbHA method in addressing the rank reversal problem, one alternative option is removed, and the ranking is recalculated with subsequent analysis of the results. The study finds that rank reversals often stem from the closeness of alternative weights. Through sensitivity analysis, this research reveals the impact of endurance attribute weight on decision outcomes, indicating that when the endurance weight reaches 0.35, the ranking of pure electric vehicles and range-extended electric vehicles reverses.
Journal Online
Links More