Intelligence Theories and Methods

381 Core Element Identification and Governance Structure Evolution of Policies: A Policy Citation Perspective Hot! Ba Zhichao, Fan Chenglei, Liu Leilei, Li Gang DOI: 10.3772/j.issn.1000-0135.2025.04.001 Identifying the core elements and structure of a policy system facilitates tracking the evolution of governance concepts and disclosing paradigms of policy formulation and design; this helps promote the reuse and migration of policy elements. This study leverages policy content mining and citation network analysis techniques to understand the evolving trajectory and governance structure of core policy objectives, instruments, and measures within a specific sector. First, a template library encapsulating the core elements of policies is developed, enabling the identification and extraction of contextual elements and citation relationships within the policy system through machine-learning algorithms. Next, a refined PageRank algorithm optimized for node jump probability is introduced to identify core elements across different periods and construct a time-series citation network by incorporating the inherent attributes and citation features of policy documents. Finally, an analysis of the dynamic ranking and evolving network structure of subcategories of policy objectives, instruments, and measures sheds light on the developmental pathways, structural configuration characteristics, and governance patterns of the core elements. Utilizing China’s large-scale energy policies from 1980 to 2020 as an empirical case, the results reveal that an optimized PageRank algorithm incorporating policy attributes provides a more precise evaluation of the actual influence and significance of policy elements. Furthermore, energy policy objectives exhibit a multi-tiered and diversified evolutionary structure, with a preferential focus on environmental policy tools, followed by demand-oriented and supply-oriented policy tools. Additionally, policy governance methods are evolving toward greater completeness and rationality, with policy changes exhibiting a compatible and incremental evolutionary trajectory. 2025 Vol. 44 (4): 381-397 [Abstract] ( 40 ) HTML (167 KB)  PDF (9791 KB)  ( 77 )

398 Theoretical Research on the Weak Signal Analysis of Disruptive Technology Based on System Thinking Hot! Tang Hulin, Su Cheng, Li Wangyu DOI: 10.3772/j.issn.1000-0135.2025.04.002 With the rapid advancement of scientific and technological revolution as well as industrial transformation, the accelerated emergence of disruptive technologies has become a general trend. The weak signals of disruptive technologies are characterized by fragmentation, ambiguity, and various forms of manifestation, posing significant challenges to understanding, analyzing, and utilizing them. Systems thinking provides a comprehensive and holistic perspective, offering the possibility of conducting an in-depth exploration of the nature and laws of the weak signal of disruptive technologies. Based on the concept of systems thinking, this study conducts a theoretical investigation of the weak signal of disruptive technologies and proposes a model for the early-stage development of disruptive technologies. The theoretical model suggests that disruptive technologies are developed in the closed-loop of technological development in the innovation chain, in which the proof-of-concept is extended in the internal elements of the development system of disruptive technologies; the effects of innovations such as methodology and technology are revealed in the proof-of-concept, the value of the technology is verified, and important signals are released to the external environmental elements; in the process of detecting weak signals, the focus should be on indicating weak signals with groundbreaking innovations. The model provides theoretical support for the early identification of disruptive technologies based on weak signals. 2025 Vol. 44 (4): 398-413 [Abstract] ( 31 ) HTML (156 KB)  PDF (1825 KB)  ( 35 )

Intelligence Technology and Application

414 Automatic Noise Reduction of Scientific Domain Document Sets Using Positive-Unlabeled Learning Hot! Chen Guo, Yang Zeyu, Chen Jing, Shao Yu DOI: 10.3772/j.issn.1000-0135.2025.04.003 In the domain analysis of science and technology, a considerable proportion of unrelated literature (impurities) exists in the datasets constructed by mainstream methods, which weakens the reliability of the final analysis results. Therefore, noise reduction is essential for removing these impurities. Performing automatic noise reduction on a dataset of domain documents without manual annotation is a prerequisite condition for whether the noise-reduction scheme can be universally applied in practice at a low cost. This study aims to transform the noise reduction task into a classification instead of a clustering problem on the premise of making full use of the characteristics of the original document dataset. We introduce positive-unlabeled (PU) learning, which can be conducted using a group of “absolutely positive samples” available in the domain dataset, to obtain reliable negative samples for the final classifiers to fit. Experiments were conducted on a dataset of journals in the MAG online library in the fields of artificial intelligence, economics, and immunology to not only compare the performance of different schemes but also construct two benchmarks and introduce Normalized Discounted Cumulative Gain as an evaluation metric, which proved the effectiveness of our method from the aspects of noise reduction revenue, usability of the result, and effectiveness of document denoising in the context of scientific and technological information analysis. 2025 Vol. 44 (4): 414-424 [Abstract] ( 21 ) HTML (119 KB)  PDF (2484 KB)  ( 23 )

425 Generative and Hierarchical Classification of Literature Based on Fine-tuned Large Language Models Hot! Hu Zhongyi, Shui Diancheng, Wu Jiang DOI: 10.3772/j.issn.1000-0135.2025.04.004 The automatic classification and indexing of literature facilitate efficient organization, storage, arrangement, and retrieval. Previous studies have primarily used discriminative models to automatically identify shallow categories of literature but have struggled with deep category classification. Hence, this study transforms the hierarchical classification problem of literature into a task of generating hierarchical category labels for literature and proposes a generative hierarchical classification indexing framework based on a large language model (LLM). The framework first uses natural language to label and interpret the hierarchical classification index of literature, then applies efficient fine-tuning techniques to perform supervised fine-tuning on the LLM. The fine-tuned LLM is then used to directly generate hierarchical classification labels for literature, and the Chinese Library Classification indices of literature are obtained via label mapping. The data from three disciplines, namely economics, medicine and health, and industrial technology, are used to evaluate the proposed model. Experimental results show that supervised fine-tuning can effectively improve the understanding and reasoning abilities of general LLMs for the classification and indexing of literature. Moreover, LLMs can achieve better classification performance than traditional discriminative models. By integrating the abstracts, titles, and keywords of literature, the classification performance of fine-tuned LLMs can be effectively improved. A comparison of Baichuan2 and Qwen1.5 models with different parameter sizes showed that the fine-tuned Qwen1.5-14B-Chat model performed the best, achieving 98% classification performance in the first level category and 80% accuracy in the most challenging fifth level category. A typical example analysis demonstrates that the fine-tuned Qwen1.5-14B-Chat has error correction capabilities. 2025 Vol. 44 (4): 425-437 [Abstract] ( 37 ) HTML (176 KB)  PDF (1606 KB)  ( 58 )

438 Knowledge Entity Extraction Method Combining Semantic Enhancement and Knowledge Distillation for Academic Literature Hot! Wang Yulong, Qin Chunxiu, Ma Xubu, Lyu Shuyue, Li Fan DOI: 10.3772/j.issn.1000-0135.2025.04.005 The accurate identification and extraction of diverse knowledge entities from large volumes of academic literature is crucial for meeting the needs of researchers and advancing fine-grained knowledge discovery. To address the issues of data sparsity and imbalances in domain-specific entities within academic literature, this study proposes an improved method that combines semantic enhancement and knowledge distillation. First, this method introduces a semantic-enhanced teacher model. By constructing an embedding representation method that integrates SciBERT, a pretrained language model based on BERT (bidirectional encoder representations from transformers), and ELMo (embeddings from language models), global semantics and dynamic word-level information are effectively combined. This approach generates more comprehensive semantic representations. Hence, it enhances the ability of the teacher model to capture complex contextual information in domain-specific academic literature. Moreover, a domain-specific pre-trained word embedding model is used to select the top n words or phrases that are most semantically related to the knowledge entities. Attention and gating mechanisms are then applied to dynamically weight the enhanced semantic information, thus effectively addressing data sparsity and the challenge of modeling long-tail entity categories. Next, a set of heterogeneous single-entity teacher models is employed to generate probability distributions across the aggregated dataset. These distributions are then used to guide the training of a student model. Finally, this study validates the effectiveness of the proposed method using three publicly available datasets from the field of materials science. Experimental results demonstrated that the proposed method achieved the highest micro F1 and macro F1 scores across three datasets in the field of materials science. Moreover, the proposed method exhibits significant robustness and generalization capabilities, particularly under scenarios of entity data sparsity and imbalance. 2025 Vol. 44 (4): 438-451 [Abstract] ( 30 ) HTML (233 KB)  PDF (1520 KB)  ( 40 )

452 Named Entity Recognition of Ancient Texts Based on the Enhancement of Multimodal Information from Chinese Characters and Pictographic Visual Alignment Hot! Zheng Xuhui, Wang Hao, Qiu Jingwen DOI: 10.3772/j.issn.1000-0135.2025.04.006 The semantic analysis and digital humanities of ancient texts are crucial for cultural development. Named entity recognition (NER) is fundamental for the subsequent knowledge discovery and organization of these texts. Therefore, developing an NER model tailored to the characteristics of simplified classical Chinese is of significant research interest. Chinese characters inherently possess substantial visual and phonetic information with pictographic features, reflecting their historical development, which can enhance entity recognition in ancient texts. This study introduces the Guwen multi-information alignment enhanced NER (GMAE-NER) model, which leverages a multimodal representation of Chinese characters. The model employs a novel approach to process and align multimodal features and integrates bidirectional encoder representations from transformers (BERT) with visual and phonetic information regarding Chinese characters, thereby improving NER performance for ancient texts. Extensive experiments on the historical text Book of the Later Han Dynasty demonstrate that the GMAE-NER outperforms baseline models, achieving a 1.32-15.00 percentage points improvement in F1 scores across various entity categories and enhancing identification of entities with overlapping expressions. Ablation studies further validate the effectiveness of the visual encoding, phonetic encoding, and feature fusion modules of the model. 2025 Vol. 44 (4): 452-465 [Abstract] ( 26 ) HTML (241 KB)  PDF (4266 KB)  ( 19 )

466 Interactive Effects of Rumor-Defying Short Videos under the Influence of Multimodal Features: Based on the Opinion Climate Mediation Perspective Hot! Fu Shaoxiong, Zeng Yuanlai, Deng Shengli DOI: 10.3772/j.issn.1000-0135.2025.04.007 The key to refuting rumors on short-video platforms is enhancing the interactive effects of these videos. Therefore, analyzing the influence of multimodal features on the interactive effects of short videos refuting rumors can provide a basis for the ecological governance improvement of short-video platform content. Based on social cognitive theory and opinion climate mediation perspective, a model of the interactive effects of rumor-defying short videos under the influence of multimodal features was constructed. A total of 2,846 effective rumor-defying short videos were captured. Using regression analysis and combining sentiment analysis, image recognition, and mediation effect tests, we explored how the publisher’s avatar, content, and title features of rumor-defying short videos affect the user opinion climate, subsequently analyzing the influence of opinion climate on the interaction effect of short videos. Avatar clarity had a significant negative effect on user opinion climate, whereas avatar authenticity, short-video theme, short-video duration, rumor-defying subjects, title length, and the emotional polarity of the title had a significant positive effect on user opinion climate. User opinion climate positively affected the interactive effects of rumor-defying short videos. Regarding the influence of external environmental factors on user behavioral responses, the user opinion atmosphere can play a fully or partially mediating role. Taking the opinion atmosphere as a mediator, this study correlates the multimodal features and interactive effects of rumor-defying short videos, elucidates the factors influencing their interactive effects, extends the research perspective on rumor-defying short videos, expands the research context of social cognitive theory, and enriches the theoretical and practical system of information behavior research. 2025 Vol. 44 (4): 466-481 [Abstract] ( 20 ) HTML (226 KB)  PDF (1474 KB)  ( 39 )

Intelligence Users and Behavior

482 Predicting Literature Download Behavior by Integrating Features of Behavioral Sequences of Academic Users Hot! Zhang Xiaojuan, Guo Jiarun, Yang Shihan, Gui Sisi DOI: 10.3772/j.issn.1000-0135.2025.04.008 In academic search systems, using historical search behavior to predict both the quantity of literature an academic user may need and the timing of that need within a given time period will help improve user satisfaction with literature recommendations. To improve the prediction accuracy of the download behaviors of academic users, the number of downloads in the next download session and the time gap until the next download session are evaluated by mining various behavioral sequence features of academic users. First, this study transforms the problem of predicting user download behavior into a time series prediction problem. Subsequently, based on the mining of the behavioral features from three perspectives, namely the query reformulation behaviors, query expressions, and download behaviors of users, the long short-term memory (LSTM) model is used to model user historical sessions as a time series to predict the download behavior. Finally, a comparative analysis of the predictive performance between the features proposed in the article and those proposed in existing research is conducted, and the predictive performances of different sets of features and individual features are explored. The features proposed in this study can improve the accuracy of prediction tasks. By clustering different users, the LSTM models that were trained on different clusters displayed the best overall predictive performance. Among all feature sets, the query expression-based features achieved the best prediction performance for the number of downloads in the next download session, and the download behavior-based features exhibited the most outstanding performance gain for the prediction of the time interval until the next download session. Owing to the limited availability of more public log datasets on academic user search behavior, this study conducted experimental validation on one dataset. Hence, the insufficient user behavior data provided by the dataset logs poses a limitation to the feature engineering approach employed in this study. 2025 Vol. 44 (4): 482-494 [Abstract] ( 28 ) HTML (243 KB)  PDF (1266 KB)  ( 48 )

Intelligence Discipline Development and Construction

495 Consolidating Disciplinary Independence of Information Science in the New Era on the Basis of Fusing the Different Research Fields Which are Both Named IS Concisely Hot! Xiao Yong DOI: 10.3772/j.issn.1000-0135.2025.04.009 Since Chinese academic circles must address their doubts about the discipline independence of information science in current and future debates, this article emphasizes that it is particularly urgent to define and consolidate its disciplinary independence in the new era to ensure its sustainable, high-quality development. This article discusses how Chinese academic circles clearly admit to the reality of different research fields coexisting in information science and sharing the same name. Next, it expounds on taking effective measures to promote the deep fusion of the two information science fields, defining their disciplinary boundaries, and making their exclusive disciplinary research distinct in the course of boosting the disciplinary construction of information science in the new era. Further, this article elaborates on the core of the exclusive disciplinary research of information science in the new era, which can be summarized as both “Running & Structure” and “Special Laws of Information Movement” in certain intelligence systems that are dominated by human intelligence, and directly inspires corresponding knowledge for specific problem solving. When all the steps are completed and the core of exclusive disciplinary research is concluded, a solid foundation for upholding the disciplinary independence of information science in the new era will be formed. The outcome of our reasoning casts no doubt on the disciplinary independence of information science in the new era. 2025 Vol. 44 (4): 495-508 [Abstract] ( 20 ) HTML (142 KB)  PDF (863 KB)  ( 34 )