Your conditions: 刘畅
  • Performance Evaluation of Chinese Universal Large Model in the Field of Humanities and Social Sciences

    Subjects: Library Science,Information Science >> Information Science submitted time 2024-05-08

    Abstract: Purpose/Significance This paper starts from the field of humanities and social sciences, and compares the model performance of humanities and social sciences from the aspects of basic knowledge and academic texts of humanities and social sciences. It aims to provide a systematic large language model evaluation benchmark for the field of humanities and social sciences for the reference of researchers in humanities and social sciences related fields. Methods/Processes Seven evaluation tasks related to the field of humanities and social sciences were designed and corresponding indicators were selected. On this basis, the current open-source and high-performance general-purpose domain Chinese large language models were selected to complete the domain-specific tasks in the form of questions and answers by invoking the local models, and their performance in the field of humanities and social sciences was quantitatively evaluated by selecting relevant indicators. Results/Conclusions The evaluation results show that among the open-source models selected in this paper, Qwen has the best performance, followed by Baichuan2, InternLM, and Atom is the worst performer in both the base model and the dialog model; moreover, in most cases, the dialog model shows more superior performance compared to the base model.

  • Research on Auxiliary Strategies of Internet Information Search Task for Elementary School Students Based on Information Need Expression Theory

    Subjects: Library Science,Information Science >> Information Retrieval submitted time 2024-04-18

    Abstract: The purpose of this study is to explore the auxiliary strategies of lower grade students in Internet information search tasks to improve their information literacy. Method/Process Based on the information needs expression theory, combined with field investigation and field interview methods, this study analyzed the process of primary school students using Baidu to search for information in the activity of Lighting up the Library . The process model of information search task is established by dividing the search task into stages, which covers the key steps such as task analysis, query construction, search result screening, content interpretation, information extraction and task completion. Result/ Conclusion The study found that lower grade students showed certain information literacy ability in information extraction and utilization, but still needed the help of auxiliary staff. In addition, the study also discusses the experience and evaluation of primary school students’ use of search engines, and finds that they have a very rich experience in using search engines such as Baidu, and give a high evaluation of the search experience of Baidu, and also raises questions from the perspective of primary school students. This study is of great significance for designing reasonable teaching activities and improving elementary school students’ information literacy, and provides a new perspective and data support for future research in related fields.

  • A Comparative Study of Search Interactions and Experiences Between Collaborative and Individual Search

    Subjects: Library Science,Information Science >> Library Science submitted time 2023-08-27 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] This study aims to explore the differences between collaborative mode and individual mode of information search on search experience and search interaction, trying to understand the characteristics of collaborative information search from various angles.[Method/process] The experiment is designed based on books interactive retrieval platform (CLEF-Social Book Search). Sixteen independent and eighteen collaborative participants are recruited to the laboratory to complete different types of book search tasks. Users' interactions with the system during search are recorded through the backend server; while their search experience is recorded by questionnaires completed before and after the search. In data analysis, users' interactions and search experience are compared between two search modes.[Result/conclusion] (1) Search experience:Compared with individual searchers, collaborative searchers have better evaluation on experiment system function and higher evaluation on aesthetics, durability, novelty of the system, and participate in the search more intently. However, collaborative searcher felt a bit more frustrated and mentally taxing than individual searchers. (2) Search strategies:Compared with individual searchers, collaborative searchers make more decisions in the goal-oriented task, especially the decision of removing books from bookbag, which indicates collaborative searchers need to sort the books in the later stage. In the open-ended task, collaborative searchers spend more time before each decision than individual searchers, which is probably because of collaborative searcher's higher level of focus and participatory. This study helps understand about collaborative search through comparison with individual search, which may provide inspiration for collaborative search system design.

  • Assessment of User Knowledge Level Before and After Searching

    Subjects: Library Science,Information Science >> Library Science submitted time 2023-08-26 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] Information search extensively influence people's daily lives, and people tend to use search engine to find the information they need. However, the existing search system is better for supporting factual tasks search, and it is not satisfactory to support tasks such as searching in order to learn. In recent years, researchers have begun to pay attention to regarding search as learning, and try to evaluate the knowledge learning in search.[Method/process] For the purpose of mastering users' learning status in search thoroughly, in this paper, 32 students from Peking University were recruited to complete user experiment. User knowledge assessment method considering knowledge quantity and quality comprehensively was established in this paper. Based on this knowledge assessment method, user knowledge level in learning tasks before and after search was evaluated respectively, and the changes of knowledge level were investigated.[Result/conclusion]The research shows that users' performance on knowledge quantity dimension becomes more comprehensive and in-depth with the search completion, and has a significant improvement in the number of knowledge points, the number of knowledge facets, the breadth of knowledge facets and the depth of knowledge facets. At the same time, there are a number of relatively professional knowledge facets after searching. Some vague knowledge concept expressions before searching were expressed more clearly and explicitly after searching. In the dimension of knowledge quality, knowledge relevance, analysis and user opinion, the vast majority of users improve after searching. This paper provides a feasible way for researchers to evaluate user knowledge learning in information search from the perspective of multi-angle evaluation of knowledge status before and after searching.

  • Research on the Knowledge Integration of Enterprises Oriented on Innovation Ecosystem

    Subjects: Library Science,Information Science >> Library Science submitted time 2023-07-26 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] Based on the innovation ecosystem scenario, this paper explores the knowledge integration of enterprises, and provides a new perspective for improving the knowledge integration effect and innovative ability of enterprises.[Method/process] Through the theoretical review of the innovation ecosystem and knowledge integration, this paper studied the process and dynamic of knowledge integration around two essential characteristics of "innovative" and "ecological", and proposed safeguard measures.[Result/conclusion] In the context of innovation ecosystem, knowledge integration of enterprises has undergone two stages:external knowledge intake and internal knowledge assimilation, and the effective integration of knowledge can be realized under the synergy of enterprises' innovative source power, industry-university-research cooperation motivation and innovative environment drive force. Furthermore, safeguard measures can be proposed for the effectiveness of knowledge integration from the three levels of innovation environment, universities and research institutes, enterprises.

  • Research on the Influencing Factors of Public Information Seeking Behavior in Major Public Health Emergencies

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] In major public health emergencies, the public has a strong demand for relevant information, and information seeking has become an important way for the public to understand and respond to risks. When facing the risk, how does the public's attitude toward information seeking shape their behaviors? What are the factors that influence the public's attitude and behavior towards information seeking? Is there any difference between different people? Answering these questions will be helpful to provide targeted information services, improve the effectiveness of individual response to the emergencies, carry out practical emergency's prevention and control solutions. [Method/process] An online questionnaire survey was conducted among 719 respondents to explore the factors affecting the public's attitude and behavior in emergency information seeking. [Result/conclusion] The research shows that there are significant differences in the attitude and seeking behaviors in different regions, ages, and educational backgrounds. Perceived risk, affective response, information sufficiency, information subjective norm and perceived behavior control positively affect the public's information seeking attitude and behavior. Meanwhile, information seeking attitude mediates between the other factors and information seeking behavior. Based on the research results, suggestions are put forward from the perspectives of information releasing channel, information communication process and information search platform.

  • Research on the Association Between Personal Academic Information Organization Behavior and Academic Procrastination of College Students

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] This paper conducts an empirical study on the association between personal academic information organization behavior and academic procrastination of college students, in order to discover the role of organizing personal academic information in preventing or correcting procrastination. [Method/process] Firstly, this paper used a questionnaire to collect data, including the Personal Academic Information Organization Level Scale for College Students, GPS scale (Chinese Revised Version), and NASP scale (Chinese Revised Version). Then, this paper used homogeneity reliability test and confirmatory factor analysis to evaluate the quality of the questionnaire. Finally, this paper used descriptive statistics, correlation test, and difference test to analyze the relationship between personal academic information organization behavior and academic procrastination of college students. [Result/conclusion] The results show that there is a significant correlation between personal academic information organization behavior and academic procrastination, and the improvement of personal academic information organization behavior may help college students to prevent or correct procrastination.

  • The Effects of Task Type and Learning Cognitive Strategies on Video-type Search Behavior in a Self-regulated Learning Context

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] In the Internet era, video platforms have become a new knowledge learning scenario. In this scenario, the search interaction process can be seen as a self-regulated learning process. This paper focuses on the effects of different task types and learning cognitive strategies on video-based search behavior in self-regulated learning contexts to complement research in the field of interactive information retrieval on video-based search behavior in learning contexts.[Method/process] In this paper, we chose two types of tasks:factual and skill-based. And cognitive strategies were classified into four types based on the self-regulated learning framework:rehearsal,elaboration,organization and critical thinking. The study adopted an experimental approach, using the Bilibili video site as a test system, and borrowed an existing measurement framework to explore the differences in behavioral characteristics and search effects of users with different learning cognitive strategies when performing different types of tasks based on experimental data.[Result/conclusion] This paper finds that different task types and learning cognitive strategies affect search behavior and learning effectiveness, showing significant differences in several indicators. The analysis and conclusion combined with task and cognitive strategy characteristics suggest improvements for the system use of video-based search platforms in learning contexts, which have positive implications for the improvement of such search platforms and the enhancement of user search experience and learning efficiency.

  • Development and Application of the Personal Academic Information Organization Level Scale for College Students

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] Taking personal information organization behavior as a whole to measure its ability and result, the paper develops the Personal Academic Information Organization Level Scale for College Students, and then applies this scale to analyze the characteristics of personal academic information organization behavior of undergraduates at Peking University to provide practical suggestions for improvement.[Method/process] In the scale development stage, the initial scale was prepared on the basis of literature review and semi-structured interview, the initial scale was purified by item analysis and factor analysis, and the quality of the final scale was evaluated by reliability test and validity test. In the scale application stage, the paper used descriptive statistics, correlation analysis, and variance analysis.[Result/conclusion] The Personal Academic Information Organization Level Scale for College Students includes five dimensions:organization level of computer documents, organization level of paper materials, academic information management literacy, cleaning level of academic space, and organization level of class notes. This scale provides a feasible tool for the measurement of personal information organization level among college students, puts forward a new perspective for the study of personal academic information organization behavior, and develops a new idea for the comprehensive understanding of information behavior of college students.

  • A Review of Research from the Perspective of Searching as Learning

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] This paper intends to sort out SAL-related studies, to form a more complete review, which provides the corresponding guidance and reference for future research in the SAL. [Method/process] In this paper, we reviewed previous studies to summarize the four main concerns in the current SAL studies:the relationship between searching and learning, the influencing factors of the search process, the evaluation of learning, and system optimization. Accordingly, a research framework for SAL, comprehensively considering context, people, system, and the relationship between searching and learning was proposed. [Result/conclusion] Future research should focus on four key issues:① characterizing learning context to design learning-related search tasks, and evaluate learning outcome corresponding to task characteristics; ② exploring how individual's learning characteristics, e.g., cognitive and metacognitive characteristics affect search interactions during the search process; ③ understanding searching as a learning process, to investigate the connections between search behaviors and learning behaviors; ④ merging contextual factors, individual factors, and searching and learning characteristics, to redesign, configure and optimize the current search system, for the integration of search system and learning system, and effectively support learning.

  • A Review of Information Behavior Studies Under Time Constraints and Time Pressure

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/Significance] Time is a universal background factor in information behaviors. However, studies on the impact of time in information behaviors from the perspective of time as a scarce resource are relatively scattered. In order to deepen the understanding of time factors and call for the attention to time in the field of information behavior, this paper summarizes the relevant research on the influence of time constraints and time pressure in the field of information behavior and summarizes the effects of time constraints and time pressure on information behaviors. [Method/Process] Based on reading and selecting related researches in the past ten years, this paper systematically summarized the effects of time constraints and time pressure on information behaviors, and focused on sorting out the roles of time constraints and time pressure in different types of information behavior processes. [Result/Conclusion] It is found that time constraints and time pressure can influence information behaviors (include searching, browsing and writing) in the process of task completion as well as information encountering without specific tasks and information decision. Future researchers should be aware of the conceptual boundaries of time constraints, time pressure and related concepts, clarify the measurement standards of time pressure, and address the positive effects of time pressure.

  • Relationships Between Note-Taking Behavior and Learning Outcome in Learning-Related Search

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/Significance] Information search is not only a process of information acquisition, but also a process of users’ self-learning. In the context of learning-related search, this paper explores the influence of the note-taking behavior on learning outcome in the process of searching, the characteristics of users’ note-taking behavior and the influence of the search context on note-taking behavior. [Method/Process] This paper collected data by the experiment method, analyzed the characteristics of note-taking by coding analysis, descriptive statistics and cluster analysis, and analyzed the association of note-taking with search context and learning outcome by difference test. [Result/Conclusion] This paper found three types of note-taking patterns: "rough collection-linear structure", "fine collection-linear structure" and "free structure". Users with more prior knowledge tend to adopt the pattern of "rough collection, linear structure", which may lead to poorer learning efficiency. The results are helpful to enrich the limited exploration of information use behavior in relevant researches on "search as learning" and inspire various information search systems to design and optimize functions from the perspective of supporting users' self-learning.

  • Research on the Effect of Prior Knowledge Differences on Collaborative Information Search and Learning

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/Significance] This study explores the differences in search interaction behaviors, collaborative experience and learning effect among groups with different levels of prior knowledge differences, so as to enrich the cognition of collaborative information search behaviors and provide suggestions for collaborative search learning in real situations. [Method/Process] The user experiment method was adopted to design two search tasks of comprehension and evaluation. Interactive behaviors were recorded by screen recording, the collaborative experience data was collected by combining questionnaires, and state and changes of knowledge before and after collaborative search were recorded by mind mapping. Statistical test methods such as variance analysis were used to analyze the effect of the data. [Result/Conclusion] The research finds: In terms of search interaction behaviors, the groups with higher prior knowledge level used more search terms than the groups with lower prior knowledge level; In terms of collaborative search experience, groups with similar prior knowledge level have higher evaluation on their own task completion; In terms of knowledge status changes, the knowledge breadth of the homogenous low level groups increase significantly more than that of the other two groups, but in terms of quality, there is little difference between the groups after search. In the end, this study puts forward some suggestions on how to improve users’ collaborative search experience and learning effect about the search system.

  • Research on the Search Behaviors Oriented to Task Difficulty Perception

    Subjects: Library Science,Information Science >> Information Science submitted time 2023-04-01 Cooperative journals: 《图书情报工作》

    Abstract: [Purpose/significance] This study uses learning-related search as the background to explore interactive behavior indicators that have a stable and significant impact on users' perception of the task difficulty. [Method/process] This research relied on course teaching to carry out behavior experiments. It took Information Technology Application course students as the research objects, used Morae to record search behavior experiments remotely. A total of 170 valid search behavior records from 39 students have been collected. According to the screen recording data and the task difficulty perception evaluation data, this study analyzed the behavior indicators which show significant differences under different difficulty levels. In addition, this study also combined two experimental data that carried out by the research team in 2016 and 2017 and the findings of existing related research, horizontal comparison was made, so as to summarize the stability of significant differences of behavior indicators under different task difficulty levels. [Result/conclusion] Finally, a stable behavior indicator system with 26 behavior indicators is obtained. The research also describes users' behavior characteristics in difficult tasks from the perspective of exploratory search process and behavior types, and points out the future research directions.

  • 用户步行导航过程中的情感变化研究

    Subjects: Library Science,Information Science >> Information Science submitted time 2017-11-08 Cooperative journals: 《数据分析与知识发现》

    Abstract:【目的】优化用户体验, 提高用户对导航系统的持续使用性。【方法】采用用户实验和基于情感词典的情感分析方法, 使用程度副词系数量、中文情感词汇本体、极性短语的强度计算公式等对情感短语进行量化, 分析影响情感的因素及其变化规律。【结果】用户在步行导航过程中的情感变化主要受到系统因素和环境因素的影响,系统更新的不及时、定位不准、环境恶劣均会对用户的情感造成影响。【局限】情感是一个复杂而且涵盖广泛的概念, 仅通过文本分析可能存在不足。【结论】通过改进系统本身和增添人性化的设置提高用户对导航系统的持续使用性, 为系统未来的升级提供有效参考。

  • 结合深度置信网络和模糊集的虚假交易识别研究

    Subjects: Library Science,Information Science >> Information Science submitted time 2017-10-11 Cooperative journals: 《数据分析与知识发现》

    Abstract: [Objective] To solve the problem of fraud transaction in e-commerce platform. [Methods] This paper proposes a method that combine Deep Belief Networks and fuzzy set based on consumers’ purchase history and reviews. Through recognizing the users in fraud transactions—cheaters to recognize the fraud transactions. [Results] Tested by experiments using the data crawled from Taobao.com, the accuracy can be achieved 89%. Compared with the shallow machine learning model, the comprehensive performance improves significantly. [Limitations] In contrast with the huge normal users and the users in fraud transactions, the experimental data in the paper is relatively small. And the test data only from Taobao.com, lack of the data from the other e-commerce platform to be validated. [Conclusions] The users in fraud transactions can be identified by the method, and the fraud transaction in e-commerce can be reduced.