基于本体的知识检索研究及实现

本文提出了一种基于本体的知识检索框架。在该框架中,知识检索主要由语义检索和基于规则的推理检索两部分组成。通过对检索方法的研究分析和算法实现,开发了一个基于本体的文档知识管理系统,并由实例对基于本体的知识检索性能予以验证。结果表明,该检索系统具有一定的智能,较好地解决了知识的重用和共享问题。     

面向监测预警过程的民族关系知识管理研究

民族关系发展演化过程中的内在规律和历史经验是监测预警的重要基础。提出基于知识管理的民族关系监测预警框架,并从知识的表达、预警过程以及知识管理实施等方面进行研究。采用民族问题案例和预警规则来表达民族关系知识,其中,案例用于组织分布广泛的难以表达的历史经验,预警规则用来表达民族关系内在规律;在预警机制实现上,将案例推理和规则推理集成在一起,能够克服各自的局限性,增强预警系统的灵活性和综合推理能力。     

(<Emphasis Type="Italic">CF</Emphasis>)<Superscript>2</Superscript> architecture: contextual collaborative filtering

Recommender systems have dramatically changed the way we consume content. Internet applications rely on these systems to help users navigate among the ever-increasing number of choices available. However, most current systems ignore the fact that user preferences can change according to context, resulting in recommendations that do not fit user interests. This research addresses these issues by proposing the \(({ CF})^2\) architecture, which uses local learning techniques to embed contextual awareness into collaborative filtering models. The proposed architecture is demonstrated on two large-scale case studies involving over 130 million and over 7 million unique samples, respectively. Results show that contextual models trained with a small fraction of the data provided similar accuracy to collaborative filtering models trained with the complete dataset. Moreover, the impact of taking into account context in real-world datasets has been demonstrated by higher accuracy of context-based models in comparison to random selection models.     

规则增强的知识图谱表示学习方法

知识图谱（Knowledge Graph,简称KG）的表示学习方法旨在将知识图谱的实体和关系表示为稠密低维实值向量, 进而在低维向量空间中高效计算实体、关系及其之间的复杂语义关联, 在知识图谱的构建、推理、融合、挖掘以及应用中具有重要作用。已有的知识图谱表示方法仅仅考虑了知识图谱中的直接事实,忽略了知识图谱中一些隐藏的语义信息,这些语义信息对于知识图谱关系和实体的嵌入表示有着重要的影响。本文提出了一种规则增强的知识图谱表示学习方法,该方法首先通过知识图谱规则挖掘的方法提取一组可代表知识图谱语义信息的Horn 逻辑规则,随后通过基于规则的物化推理方法将相应的隐藏语义信息注入到知识图谱表示学习模型中。实验结果表明,基于规则增强的方法可以显著提升已有知识图谱表示学习模型在链接预测和定理预测上的效果和性能。     

面向专利领域的汉英机器翻译融合系统

面向专利领域的机器翻译近年来已成为机器翻译的重要应用领域之一。本文提出了一个汉英专利文本机器翻译融合系统,该系统以规则系统为主导搭建,并把规则翻译方法和基于短语的统计翻译系统相结合。在融合系统中,规则系统主要负责源语言的分析和转换阶段的处理,生成相应的源语言句法分析树与转换树,并确定目标语言的基本句法框架。统计翻译系统则在目标语生成阶段根据生成的目标语句法结构寻找合适的对译词形,并产生最终的候选译文。通过利用自动评测指标对融合系统进行测试,融合系统的结果均优于单个规则系统和统计系统的结果,表明了融合方法的有效性和可行性,可以改善系统的翻译性能,提高翻译质量。     

新闻推理与论证的语用逻辑研究

抽离语用属性、高度形式化的新闻推理与论证的分析及其评价方式已不能准确刻画、涵盖新闻报道中大量运用的、真实的推理与论证模式。当下非形式逻辑理论正在扩充和修改新闻推理与论证的既有形式化内涵,在并不排斥形式逻辑理性的前提下,致力于发现、分析和发展传受双方在日常新闻报道中所运用的论证标准、程序和模式。这些新闻推理与论证逻辑的语用价值走向对于新闻思维的具体内容和逻辑形式关系的建构有相当的现实意义。     

Comparison of human performance by knowledge domain: Types, frequency, and sequencing of errors made while interacting with an information system

This study investigated whether measures of error type, frequency, and sequencing contribute to understanding how different types of domain knowledge influence human performance when using information systems. Micro-level analyses of errors were conducted to compare the performance of two groups using a geographical information system (GIS). Error types, error frequency, error sequencing, and task completion were measured and compared for the two groups. The two groups were similar in frequency of type of errors; both committed terminological and trial errors more frequently than the other 10 types. Error patterns or sequences were likewise similar, and domain of expertise did not affect task completion time. The group with domain knowledge in computer science showed a somewhat greater tendency to commit certain knowledge-based errors (KBEs), indicating an association between knowledge-based errors and procedural knowledge. The group with domain knowledge in geography committed certain rule-based errors (RBEs) more often, indicating an association between rule-based errors and declarative knowledge. Therefore, error type does hold promise as a measure of the relationship between domain knowledge type and human performance using information systems.     

分布式环保档案信息资源共享系统研究(续)

本文是上期同名文章的续篇,提出了分布式环保档案信息资源共享系统的总体架构、功能体系及部署应用模式。分布式环保档案信息资源共享系统包括后台著录管理和前台共享服务两个子系统,由部署在各级环境保护行政主管部门的节点共同组成,不仅可以实现各单位环保档案信息资源的自治管理,而且通过网络服务互操作,可实现部门间环保档案信息资源的安全交换和分级共享。     

Knowledge bases in medicine: a review

Efforts to represent knowledge effectively have been central to progress in various aspects of medical informatics. These efforts range from relatively simple "electronic textbooks" to fairly sophisticated knowledge-based systems, which function as well as, or even better than, human experts faced with similar problems. Knowledge bases have been developed in many fields, but the relatively limited domains and structured language of medicine, as well as the importance of information in the provision of good medical care, have made research in medical knowledge representation an area of intense activity. This paper reviews representative knowledge bases and knowledge-based systems in medicine: electronic textbooks such as PDQ and the Hepatitis Knowledge Base (HKB), rule-based systems such as MYCIN, causal models (e.g., CASNET), and hypothesis- or frame-based systems, exemplified by PIP and INTERNIST-1. The paper describes the relationships among divergent approaches and provides a sense of current and future trends. It examines problems in knowledge-based systems, particularly in knowledge representation and acquisition, and the responses to these challenges. The latter include the use of domain-independent software shells for constructing knowledge bases, the adaptation and use of previously existing knowledge bases, and multiple uses of the same knowledge base for different purposes.     

Simulation of the Reference Process,Part II:

This paper is a continuation of a previous one in The Reference Librarian, exploring the advantages of a simulation (rather than expert system) approach to designing a system to relieve reference desk overload. Another paper examined the correlation between four expert systems and various reference models; it made several recommendations for future research. The ideas presented in those two earlier papers have now been implemented in a prototype system named REFSIM, which can be used by either client or librarian. REFSIM can simulate a reference librarian, a teacher of reference librarians and a client, singly or in combination. It has an expert-system consultation mode and several Intelligent Computer-Assisted Instruction (ICAI) modes. The components and architecture of REFSIM are described in this paper. Details, including a sample transaction, are given for the expert system mode, which is to be used for consultation by library clients. An overview is supplied for two ICAI modes, which can be used to instruct clients or train reference librarians.     

Florida's marine resource information system: A geographic decision support system

Decision support systems (DSS) are a class of information systems where data, models, and an interface are combined to support a decision-maker's needs for data and analysis. This article reports on a system which fits all of the classic definitions of a DSS and which includes spatial models of object locations as crucial parts of the analytic, data, and interface support provided to the user. The system illustrates several interesting aspects of the construction of systems of this type, including the potential role of geographic information system (GIS) capabilities in DSS; the translation of user decision support needs with a geographic component into a DSS architecture; and the integration of a PC-based GIS package with additional interface, data management, and analytic tools. The system also illustrates certain managerial implications for systems of this type, including the importance of planning for system maintenance and the value of geographic data.     

Development aid programs for bit-slice based systems

Though bipolar or bit-slice microprocessors offer a considerable advantage in execution speed and processing capability, it is generally very hard to get them operational. The difficulty usually resides in the complexity of the microinstructions (several tons of bits) and the intimate relationship between them and the hardware for which they are intended. This report describes a set of programs designed to facilitate microprogramming in the case of systems based on AM2901 units. The set consists of a hardware specification program, an outer program for microprograms written in an ad hoc symbolic language, a program to simulate the execution of symbolic microprograms, an assembly program and an optimisation program. The use of these programs has made it possible to obtain a considerable reduction in the cost of multiprogrammed systems, at both development and maintenance level.     

Distributed knowledge acquisition for the automatic construction of integrated expert systems

The authors discuss the problem of distributed knowledge acquisition for the construction of complete and consistent databases in integrated expert systems via the sharing of knowledge sources of different topologies (experts, problem-oriented texts, and electronic media in the form of databases). The emphasis is on the models, methods, and algorithms of distributed knowledge acquisition from databases as additional knowledge sources. The authors describe the architecture and basic facilities of distributed knowledge acquisition, which function as a part of the AT-TECHNOLOGY tool complex.     

Modeling of intelligent agent interactions for multiagent systems

The experience gained in the development of the system of simulation modeling of intelligent agent interaction (SMIAI) is described. It is based on the general model of agent interaction proposed by G.V. Rybina, who developed the works on intelligent dialog systems, including the well-known Popov’s communication model, in the late 1990s. The goal of SMIAI is to validate the applicability of the proposed models, methods, and algorithms of intelligent agent interactions and their architecture in developing MAS prototypes for different problem areas. The SMIAI architecture, support means for simulation experiments, transmission and processing of messages, and other components are considered. Realized MAS prototypes for several applications are presented as examples. The models of local, thematic, and global dialog structure for intelligent agents are presented and the problems of intelligent agent communications behavior formation are analyzed. The creation of the models under consideration by means of G2 and Microsoft Visual Studio is described.     

基于知识空间的智能信息检索模型研究

对当前信息检索系统存在的问题进行分析，明确智能信息检索迫切需要解决的问题；在对现有的智能信息检索模型比较分析的基础上，提出基于知识空间的智能信息检索模型——知识空间模型，并描述基于知识空间的智能信息检索模型的框架、体系结构及其中若干关键问题。     

基于中文学术文献的领域本体概念层次关系抽取研究

基于学术文献构建领域本体对促进领域学科发展具有重要的意义。本文提出了一种以中文学术文献为数据源,半自动化抽取领域本体层次关系的框架方法。首先,构建了一个通用的领域本体层次关系的细粒度研究框架。其次,设计了一种新的概念表示方法,融合了深度学习方法得到的概念语义特征和上下文的时间序列词频。进一步结合了AP聚类、Prim算法和Web搜索引擎的查询数据,提出了基于规则推理的本体概念层次关系抽取算法(RROCHE),实现了半自动化概念层次关系抽取。最后,基于中文分词领域的中文学术文献数据,通过数值实验方法讨论了方法的可行性和有效性。本文提出的框架方法也非常容易推广并应用到各领域本体层次关系任务中。     

Flagging fake news on social media: An experimental study of media consumers' identification of fake news

Policymakers are taking action to protect their citizens and democratic systems from online misinformation. However, media consumers usually have a hard time differentiating misinformation from authentic information. There are two explanations for this difficulty, namely lazy reasoning and motivated reasoning. While lazy reasoning suggests that people may feel reluctant to conduct critical reasoning when consuming online information, the motivated reasoning theory points out that individuals are also thinking in alignment with their identities and established viewpoints. A proposed approach to address this issue is adding fact-checking flags in the hope that flags could alert people to information falsehoods and stimulate critical thinking. This study examines the impact of fact-checking flags on media consumers' identification of fake news. Conducting an experiment (n = 717) on Amazon Mechanical Turk, the study finds that experimental participants with different political backgrounds depend heavily on flag-checking results provided by flags. Flags are powerful to influence people's judgments in a way that participants have blind beliefs in flags even if the flag assessments are inaccurate. Furthermore, the study's results indicate that flag assessments made by professional fact-checkers or crowdsourcing are equally influential in shaping participants' identification. These observations provide public and private leaders with suggestions that fact-checking flags can significantly affect media consumers' identification of fake news. However, flags appear to have little ability to promote critical thinking in this experiment.     

组织知识创新能力开发的内在心智基础及演化过程分析

组织知识创新能力是指组织作为一个整体通过对各类与组织发展相关的新知识的发现与发明,从而创造产生出新的组织知识的能力。在组织知识创新能力开发过程中,组织共享心智模型既为组织中各种知识的复杂协作与一体化提供着潜在的认知背景,又为组织知识载体———知识员工的创造力的激发提供着内在的精神意义支持。本文正是基于这一假设出发,按照组织知识创新流程的不同阶段,对组织共享心智模型在组织知识创新能力开发过程中的作用进行了系统的分析,最后给出一个拓扑算子模型进行了数理实证。     

Television use,stereotypes of African Americans and opinions on affirmative action: An affective model of policy reasoning

This paper proposes and tests an affective model of policy reasoning with television use as a first stage variable in a heuristic causal chain. We predict that TV use leads to affect which then leads to opinions on related issues. We test the power of the model to explain how citizens make up their minds regarding government policies on affirmative action, with stereotypes of African Americans as the second stage, affective variable. Data from White college students indicate that there was good fit (non‐significant chi‐squares) for the model. The paths explaining the most variance were those from perceived negative TV attributes to stereotypes, and stereotypes to opinions. Perceptions of negative TV portrayals predicted negative stereotypes, which predicted opposition to affirmative action policies. Implications for media use and heuristic reasoning models are discussed.     

Understanding the relationships between information architectures and business models: An empirical study on the success configurations of smart communities

With the development of Internet of Things (IoT) and big data, many smart city and smart community projects bloomed in recent years. Following the two approaches of smart city development from Kuk and Janssen (2011), the study proposed a synergetic framework for understanding the relationship between information architectures and business models. Since community is a basic unit of a city, the development goals of a smart city are needed to be implemented at the level of communities. The development path of smart community is a configuration set including both information architecture factors and business model patterns. Based on the cases of 69 communities from Beijing, China, we explored successful configurations based on the framework. Using the Qualitative Comparative Analysis (QCA) method, we found that a successful smart community depends on the integration between information architectures and business models, and different business models rely on different information architectures elements. Networking, terminals, and sensors are key information architecture elements that are used more frequently in business models.