Iterative rule-guided reasoning over sparse knowledge graphs with deep reinforcement learning

In recent years, reasoning over knowledge graphs (KGs) has been widely adapted to empower retrieval systems, recommender systems, and question answering systems, generating a surge in research interest. Recently developed reasoning methods usually suffer from poor performance when applied to incomplete or sparse KGs, due to the lack of evidential paths that can reach target entities. To solve this problem, we propose a hybrid multi-hop reasoning model with reinforcement learning (RL) called SparKGR, which implements dynamic path completion and iterative rule guidance strategies to increase reasoning performance over sparse KGs. Firstly, the model dynamically completes the missing paths using rule guidance to augment the action space for the RL agent; this strategy effectively reduces the sparsity of KGs, thus increasing path search efficiency. Secondly, an iterative optimization of rule induction and fact inference is designed to incorporate global information from KGs to guide the RL agent exploration; this optimization iteratively improves overall training performance. We further evaluated the SparKGR model through different tasks on five real world datasets extracted from Freebase, Wikidata and NELL. The experimental results indicate that SparKGR outperforms state-of-the-art baseline models without losing interpretability.     

集群企业网络嵌入对技术创新的影响——基于知识的视角

通过文献回顾,构建了集群企业网络嵌入、知识转移、知识创造与技术创新之间关系的理论模型,并以浙江杭州软件产业集群和文化创意产业集群企业为调研对象,对模型进行了检验。研究结果发现,集群企业网络嵌入对知识转移具有显著正向影响;集群企业网络嵌入对知识创造具有显著正向影响;集群企业网络嵌入并不直接影响技术创新,而是通过知识转移与知识创造间接影响技术创新;集群企业知识转移并不直接影响技术创新,而是通过知识创造的完全中介作用间接影响技术创新;加入高新技术集群类型这一控制变量后,各变量间影响关系的检验结果存在一定差异。     

A neural knowledge graph evaluator: Combining structural and semantic evidence of knowledge graphs for predicting supportive knowledge in scientific QA

Effectively detecting supportive knowledge of answers is a fundamental step towards automated question answering. While pre-trained semantic vectors for texts have enabled semantic computation for background-answer pairs, they are limited in representing structured knowledge relevant for question answering. Recent studies have shown interests in enrolling structured knowledge graphs for text processing, however, their focus was more on semantics than on graph structure. This study, by contrast, takes a special interest in exploring the structural patterns of knowledge graphs. Inspired by human cognitive processes, we propose novel methods of feature extraction for capturing the local and global structural information of knowledge graphs. These features not only exhibit good indicative power, but can also facilitate text analysis with explainable meanings. Moreover, aiming to better combine structural and semantic evidence for prediction, we propose a Neural Knowledge Graph Evaluator (NKGE) which showed superior performance over existing methods. Our contributions include a novel set of interpretable structural features and the effective NKGE for compatibility evaluation between knowledge graphs. The methods of feature extraction and the structural patterns indicated by the features may also provide insights for related studies in computational modeling and processing of knowledge.     

跨国公司母子公司关系研究-基于知识的视角

摘要：基于知识的跨国公司母子公司关系是跨国公司母子公司关系的重要形式之一。本文主要从知识的视角对跨国公司母子公司的关系进行研究。分析了跨国公司母子公司间的知识构成与特性,并对跨国公司母子公司双方知识投入对母子公司关系的形成的影响进行了研究,在对母子公司间的知识转移研究的基础上,得出结论：跨国公司母子公司关系会随母子公司间知识转移而发生变化,子公司拥有当地化知识作用的提高使得跨国公司子公司在母子公司关系中越来越占据主导地位。     

Mining multigranularity decision rules of concept cognition for knowledge graphs based on three-way decision

Machine understanding and thinking require prior knowledge consisting of explicit and implicit knowledge. The current knowledge base contains various explicit knowledge but not implicit knowledge. As part of implicit knowledge, the typical characteristics of the things referred to by the concept are available by concept cognition for knowledge graphs. Therefore, this paper attempts to realize concept cognition for knowledge graphs from the perspective of mining multigranularity decision rules. Specifically, (1) we propose a novel multigranularity three-way decision model that merges the ideas of multigranularity (i.e., from coarse granularity to fine granularity) and three-way decision (i.e., acceptance, rejection, and deferred decision). (2) Based on the multigranularity three-way decision model, an algorithm for mining multigranularity decision rules is proposed. (3) The monotonicity of positive or negative granule space ensured that the positive (or negative) granule space from coarser granularity does not need to participate in the three-classification process at a finer granularity, which accelerates the process of mining multigranularity decision rules. Moreover, the experimental results show that the multigranularity decision rule is better than the two-way decision rule, frequent decision rule and single granularity decision rule, and the monotonicity of positive or negative granule space can accelerate the process of mining multigranularity decision rules.     

HNERec: Scientific collaborator recommendation model based on heterogeneous network embedding

Scientific collaboration contributes to the innovation of science and technology. For individual scholars, it is valuable to find potential and valuable collaborators to improve the quality and quantity of research. Previous research mainly recommends collaborators by exploring one kind of relationship between authors, such as the research topics or co-authorship. Some research tries to utilize more relationships, but they simply integrate the recommendation results of different relationships rather than integrating them. This study proposes a novel Heterogeneous Network Embedding Recommendation (HNERec) model for scientific collaborator recommendation. The HNERec model employs a random walk to develop a heterogeneous network considering four kinds of meta-path illustrating the topic relationship, authorship, citation relationship, and venue relationship. It utilizes the skip-gram model to embed the nodes, and finally generates a recommendation list based on the similarity between the corresponding node vectors. Experiments on a DBLP dataset and an AMiner dataset demonstrate the reliability of the HNERec model. The HNERec model outperforms the mainstream collaborator recommendation baselines in metrics such as F1, MRR, and nDCG. The results can not only be applied to the selection of collaborators by individual scholars but also allow scientific and conference institutions to recommend collaborators.     

PVE: A log parsing method based on VAE using embedding vectors

Log parsing is a critical task that converts unstructured raw logs into structured data for downstream tasks. Existing methods often rely on manual string-matching rules to extract template tokens, leading to lower adaptability on different log datasets. To address this issue, we propose an automated log parsing method, PVE, which leverages Variational Auto-Encoder (VAE) to build a semi-supervised model for categorizing log tokens. Inspired by the observation that log template tokens often consist of words, we choose common words and their combinations to serve as training data to enhance the diversity of structure features of template tokens. Specifically, PVE constructs two types of embedding vectors, the sum embedding and the n-gram embedding, for each word and word combination. The structure features of template tokens can be learned by training VAE on these embeddings. PVE categorizes a token as a template token if it is similar to the training data when log parsing. To improve efficiency, we use the average similarity between token embedding and VAE samples to determine the token type, rather than the reconstruction error. Evaluations on 16 real-world log datasets demonstrate that our method has an average accuracy of 0.878, which outperforms comparison methods in terms of parsing accuracy and adaptability.     

知识转移阻隔解析——从知识本体的视角出发

知识转移阻隔现象不仅源于客观环境,知识本体也是一个重要原因.从本体视角出发,从知识的分类、知识的层级分布、知识的特征,以及知识的工具价值等四个角度,解释了知识转移阻隔的形成机理;分析了四个角度之间的关系,讨论了理论对管理实践的启示.     

产业集群知识创新机制的一个新视角——基于知识发酵理论研究

我国许多地方通过产业集群提高了产业竞争优势,拉动了地方经济的增长。产业集群拥有强大的知识创新能力,因此研究产业集群知识创新机制对于提高产业竞争力和促进区域经济的发展具有重要的理论意义和现实意义。文章借助仿生学原理,运用知识发酵理论阐述了产业集群知识创新SEC I过程各要素构成,弥补了当前有关知识创新理论的不足,为产业集群知识创新提供了一种新的思路和理论框架。     

From free to paid: Effect of knowledge differentiation on market performance of paid knowledge products

Exploring the factors that affect the market performance of paid knowledge products is of great importance for knowledge payment platforms. Drawing on the sensations-familiarity framework and social capital theory, this study investigates how knowledge differentiation between paid and free knowledge impacts market performance, along with the moderating effect of knowledge providers’ social capital. Technically, a neural network-based text mining model is utilized to transform free and paid knowledge to semantic vectors, whose dissimilarity is calculated as knowledge differentiation. Empirical analysis on a real dataset reveals the positive (or negative) effect of knowledge differentiation on sales (or eWOM, electronic word of mouth), which will be more prominent with the increase of social capital. The results are reinforced with robustness checks regarding alternative knowledge-differentiation measures, more control variables and alternative regression methods. The present study extends our understanding of knowledge payment and free-to-paid consumption, and offers practical implications for content design and product management.     

遵从科创规律:再谈科创成果与年龄的关系

本文回顾了"成果-年龄"研究的历史和逻辑,系统梳理了相关研究,讨论了该理论模型存在的问题和下一步发展方向.认为"成果-年龄"理论模型在宏观和中观层次上具有一定同构性,可以借鉴"介尺度"思想方法,将威布尔分布统计规律在微观层次尝试.建议要遵从科创规律,保障科创群体工作时间和专注力,在最适合出成果的年龄专心搞研究.     

动态联盟：企业提高竞争力的有效手段

企业动态联盟是随着信息技术的发展和经济全球化而出现的一种组织运作模式，它也是企业获取竞争优势、实现快速增长的重要战略。本文论述了企业动态联盟的基本理论，详细分析了动态联盟对企业提高竞争力的重要作用、动态联盟的有效运作，最后探讨了动态联盟在我国企业的实践与应用。     

Learning capability: the effect of existing knowledge on learning

It has been observed that different people learn the same things in different ways – increasing their knowledge of the subject/domain uniquely. One plausible reason for this disparity in learning is the difference in the existing personal knowledge held in the particular area in which the knowledge increase happens. To understand this further, in this paper knowledge is modelled as a ‘system of cognitive schemata’, and knowledge increase as a process in this system; the effect of existing personal knowledge on knowledge increase is ‘the Learning Capability’. Learning Capability is obtained in form of a function, although it is merely a representation making use of mathematical symbolism, not a calculable entity. The examination of the function tells us about the nature of learning capability. However, existing knowledge is only one factor affecting knowledge increase and thus one component of a more general model, which might additionally include talent, learning willingness, and attention.     

遵循生化返规律开发儿童智力的实验

遵循生化规律能否开发２岁儿童智力？经过１ａ的实验，结果表明，实验组２９名儿童平均智商为１２８．１４，与对照组２３名儿童平均智商１０３．９１相比较，具有显著性差异。     

企业多元化的产学研知识互动形式——基于因子分析的实证研究

为探索企业参与多元化产学研知识互动的情况,对广东省清溪镇所有规模以上企业进行调研,获得166家企业参与14种产学研知识互动情况的数据。因子分析结果显示,企业多元化产学研知识互动形式可分为合作研究、人员流动、研究服务与支持和科研成果商业化四类。其中研究服务与支持形式的知识互动程度最高,其次是合作研究和人员流动,而科研成果商业化的最低。但总体上企业参与产学研知识互动的程度偏低。     

运用合理情绪疗法改善高中生人际交往障碍的个案研究

本案例运用合理情绪疗法为一名因人际交往出现情绪困扰的高中生来访者进行了心理咨询。针对其不合理的信念和认知模式采取认知行为矫正,引导其进行认知训练以形成合理的理性认知,从而缓解不良情绪,最终达到咨询效果。     

使用调度和管制机制改善网络QoS的研究

本文从改进网络的“尽最大努力交付”角度,详细阐述了调度和管制机制在改善网络的QoS方面的原理和应用,并给出了理论上的计算公式。     

运用多媒体提高高校教学效率研究

本文探讨了在信息化时代,高校教师如何充分利用多媒体去提高大学教学的效率。学习活动不只是以教师为中心获取信息,而应该引起学生们的浓厚兴趣。本文结合了自己的教学体验,阐述了大学英语教师和学生在信息化的教学环境下各自的角色和作用。     

运用合理情绪疗法改善高中生人际交往障碍的个案研究

本案例运用合理情绪疗法为一名因人际交往出现情绪困扰的高中生来访者进行了心理咨询。针对其不合理的信念和认知模式采取认知行为矫正,引导其进行认知训练以形成合理的理性认知,从而缓解不良情绪,最终达到咨询效果。     

Relational effects on knowledge integration: the differential effects on search and transfer

The paper investigates the effect of homophily and status differences on knowledge integration among individuals involved in product development activities. By distinguishing between search and transfer as two integral parts of the knowledge integration process, the paper addresses the question of how cognitive and motivational barriers to knowledge integration operate differently through the search and the transfer processes respectively. A study of 434 knowledge integration activities among 49 participants in four software development teams suggests that while search is directed both by homophily and by status differences among team participants, only status differences affect the perceived value of knowledge transfer activities. Further, the study indicates that there is not necessarily a correspondence between the status cues that direct search activities and the status cues that drive the perceived value of knowledge transfer activities among team participants.