一种基于图卷积自编码模型的多维度学科知识网络融合方法     

李慧  胡吉霞 《图书情报工作》2020,64(18):114-125

[目的/意义] 针对包含单一类型知识单元的知识网络难以全面反映学科知识结构的问题，提出一种从多维度进行知识网络结构融合的方法，为学科领域知识结构挖掘提供借鉴。[方法/过程] 利用LDA及TF-IDF方法抽取学科知识单元，然后运用语义相似度和关键词共现分析方法构建3个学科知识子网络：主题网络、关键词网络和实体网络，并采用空间节点传递对齐方法对齐子网络节点，接着设计基于图卷积操作的自编码模型对知识节点进行表示，最后通过计算余弦相似度重构学科知识网络。[结果/结论] 实验部分以人工智能领域为例，构建融合主题、关键词和实体的学科知识网络并展开分析，实验结果表明，本文所提方法能有效地揭示学科领域研究内容和知识结构，为学科知识发现与组织研究提供有益参考。  相似文献   

对2019年中国电视剧研究的分析与反思     

金丹元  陈莎 《美育学刊》2020,11(2):47-54

2019年正值新中国成立70周年,我国电视剧创作出产了许多革命历史题材和现实主义电视剧佳作,同时网络剧也开始从边缘走向中心,成为观众喜闻乐见的电视剧形式。通过梳理、概括国内重要的学术期刊、报纸和会议论文中关于电视剧研究的文章,并从内容创作、美学价值、类型特点和女性塑造等几个方面进行解读,可以归纳出本年度电视剧研究的热点和焦点,同时也提出了对电视剧研究新动向的一些看法,以期为电视剧研究和创作提供借鉴。  相似文献   

知识转移视角下复杂产品产学研协同创新管理机制研究     

李培哲  菅利荣  刘勇 《科技管理研究》2019,39(2)

复杂产品协同创新是一项复杂的系统工程,具有研发周期长、系统组成复杂、技术难度和投资风险大等特点,传统管理模式很难满足其产学研协同创新的要求,需要科学、合理的管理机制作为保障。以知识转移为视角,从关系契约管理和非正式网络管理两方面设计了复杂产品产学研协同创新的管理机制,提出关系契约管理和非正式网络管理在复杂产品产学研创新活动中会协同发生作用,能有效促进创新主体间显性知识和隐性知识的转移,保证产学研协同创新活动的顺利进行。  相似文献   

我国体育彩票销售额的空间异质性及影响因素分析          下载免费PDF全文

王成龙 《河北体育学院学报》2019,33(6):41-48

收集2013—2017年我国31个省份体育彩票销售额的面板数据和2017年各省份社会发展指标的相关数据，运用探索性空间数据分析方法（ESDA），对31个省份体育彩票销售额的空间异质性特征及影响因素进行研究。全局空间自相关分析结果表明，体育彩票销售额具有明显的空间聚集性，且呈显著的空间正相关性。经分析，发现31个省份体育彩票销售额不均衡，东部沿海省份的销售额高于西部省份，提出应重点关注西部地区的体育彩票事业发展。31个省份体育彩票销售呈现出以次冷区域为主，个别热点区域和次热区域相结合的特点，说明我国体育彩票销售的区域性寡占市场特征较明显，提出应加强省份间的沟通交流，促进各区域联动均衡发展。在影响因素方面，人均GDP、体育场地面积对体育彩票销售额具有显著性影响，其中人均GDP对南方的影响大于北方，体育场地面积对东北地区的影响大于西南地区，提出应进一步加大对体育场地建设的投入力度，以促进我国体育彩票事业的发展。  相似文献   

因果要素关联视域下社交媒体倦怠用户画像模型构建   总被引：1，自引：1，他引：0  

张艳丰  彭丽徽  洪闯  刘金承  高凯 《图书情报工作》2019,63(14):94-100

[目的/意义]探究社交媒体倦怠的原因要素和结果要素标签，分析各要素间的相互作用关系，扩展用户信息行为的理论研究范畴，为企业深入了解用户社交媒体倦怠发展态势提供指导。[方法/过程]基于信息生态理论，提取"信息人、信息、信息环境和信息技术"4个信息生态因子维度的社交媒体用户画像标签；根据用户心理特征和行为表现，抽取用户画像情感和行为标签，构建社交媒体倦怠用户画像模型。[结果/结论]透过社交媒体用户消极使用和负面情感变化的典型社会现象，对用户画像模型进行整合分析与关系凝练，证明本研究社交媒体倦怠用户画像模型与S-S-O理论框架具有一致性，并根据实际进行理论上的拓展和应用。  相似文献   

基于舆情大数据的社会安全事件情报感知与应用研究     

夏一雪 《现代情报》2019,39(11):121-127

[目的/意义] 基于舆情大数据进行社会安全事件情报感知，为社会安全事件应急管理提供情报支持。[方法/过程] 基于舆情大数据的社会网络属性，剖析网络社会环境下社会安全事件特征和情报感知需求，提出基于舆情大数据进行社会安全事件情报感知具有多元、快速、全面、深度的优势。从社会安全事件要素和过程两个方面，构建全要素、全周期的情报感知指标体系。分析基于移动传感网络的情报感知路径，针对情报感知指标进行情报感知，确定"异常→风险→情报"的情报感知过程，并研究情报感知方法。[结论/结果] 基于舆情大数据构建情报感知指标体系并开展应用研究，为网络社会环境下社会安全事件应急管理提供新思路和新方法。  相似文献   

基于深度学习的MOOC论坛探索型对话识别方法研究     

董庆兴  李华阳  曹高辉  夏立新 《图书情报工作》2019,63(5):92-99

[目的/意义]大规模在线开放课程论坛具有丰富的用户评论数据。从大量未区分的评论数据中，自动识别出知识密度较高的探索型对话并挖掘其潜在价值，对于改善教师教学质量以及提高学生知识水平具有重要影响。[方法/过程]首先利用GloVe方法训练词向量，加强对文本语义的理解，然后利用卷积神经网络自动学习文本特征，提出一种基于深度学习的探索型对话自动识别模型，并在学堂在线平台《心理学概论》课程论坛标注数据集上进行实证与对比研究。[结果/结论]实验结果显示，利用GloVe方法预训练词向量以及在训练过程中不断对词向量进行学习修正能够提高模型效果。该模型识别探索型对话的F1值为0.94，相较于传统的朴素贝叶斯方法（0.88）、逻辑斯谛回归方法（0.89）、决策树方法（0.88）以及随机森林方法（0.88）取得较大提升，具有较高的实用性和较低的学习成本。  相似文献   

Network ethnography applied: understanding the evolving health and physical education knowledge landscape     

Leigh Sperka  Eimear Enright 《Sport, Education and Society》2019,24(2):168-181

The progressively global, neoliberal, privatised, and digital education environment poses new methodological challenges for educational researchers, prompting a need to innovate. It has been suggested, however, that better commentary and reflection on methodological innovation in education is required. This paper considers the benefits and challenges associated with the use of network ethnography, one methodological approach that has emerged to address new social complexities. We explain the rationale for, and process of, this methodology through reference to an illustrative case: a network ethnography of the outsourcing of Health and Physical Education (HPE) curricular work to external providers. In doing this we map and critique three interrelated activities (i.e. Internet searches, interviews, and network diagram construction) that constituted our network ethnography. The discussion turns to how network ethnography allowed us to access new knowledge about the outsourcing of HPE curricular work to external providers by, for example, facilitating us in asking different questions, and foregrounding various stakeholders, networks and relationships that we may not have discovered had we relied on other approaches. This illustrative case demonstrates the capacity of network ethnography to generate rich data and offers a provocation for educational researchers to consider expanding their methodological repertoire.  相似文献   

Methods for assessing higher education research team collaboration: comparing research outputs and participant perceptions across four collaborative research teams     

Nick Kelly  Joanne Doyle  Marisa Parker 《高等教育研究与发展》2020,39(2):215-229

ABSTRACT

This article suggests that situations in which multiple research teams are convened under similar conditions present an opportunity to discover factors that lead to productive collaboration. It argues that social network analysis of research team outputs becomes more valuable when paired with data about research participant perceptions; and that any variables used as indicators of collaboration need to be calibrated using datasets from multiple studies with cross-team comparisons. The article provides an example of the kind of methodology needed to achieve this, describing a study with data from four research teams based at an Australian university campus, reporting their research performance over four years under conditions in which many variables were controlled and with results augmented by a survey of participant perceptions. Findings from the study indicate that there were exceptions to hypothesized associations between participant perceptions of collaboration and specific social network analysis measures over co-authorship data. The article suggests that, given the methodological challenges of studying research teams in the field, multiple datasets combining findings such as those in the present study are a path towards the development of indicators of productive higher education research collaboration.  相似文献   

Knowledge-guided multi-granularity GCN for ABSA     

《Information processing & management》2023,60(2):103223

Aspect-based sentiment analysis aims to determine sentiment polarities toward specific aspect terms within the same sentence or document. Most recent studies adopted attention-based neural network models to implicitly connect aspect terms with context words. However, these studies were limited by insufficient interaction between aspect terms and opinion words, leading to poor performance on robustness test sets. In addition, we have found that robustness test sets create new sentences that interfere with the original information of a sentence, which often makes the text too long and leads to the problem of long-distance dependence. Simultaneously, these new sentences produce more non-target aspect terms, misleading the model because of the lack of relevant knowledge guidance. This study proposes a knowledge guided multi-granularity graph convolutional neural network (KMGCN) to solve these problems. The multi-granularity attention mechanism is designed to enhance the interaction between aspect terms and opinion words. To address the long-distance dependence, KMGCN uses a graph convolutional network that relies on a semantic map based on fine-tuning pre-trained models. In particular, KMGCN uses a mask mechanism guided by conceptual knowledge to encounter more aspect terms (including target and non-target aspect terms). Experiments are conducted on 12 SemEval-2014 variant benchmarking datasets, and the results demonstrated the effectiveness of the proposed framework.  相似文献