新闻事件检测是自然语言处理任务中的一项任务.新闻事件检测旨在从新闻文本数据流中检测出新闻事件并给出事件主题.人工构建新闻事件的特征费时费力.传统的新闻事件检测方法是根据新闻事件之间的空间距离检测新闻事件,对于不同的新闻事件相似度较高时,容易误判为同一事件.针对上述问题,论文提出基于注意力机制的双向长短记忆网络构建新闻事件检测模型,通过深度学习学习新闻文本深层次的特征并且基于新闻事件检测模型构建新闻事件建模应用系统.实验表明论文方法在准确率、召回率优于传统方法,可对新闻事件准确识别.

In order to solve the problem that traditional word vectors are difficult to express the context semantics and the feature extraction of traditional model is single, a multi-feature fusion model named BERT-BiLSTM-IDCNN-Attention-CRF for Named Entity Recognition is proposed, which uses BERT to model the context semantic relationship of word vectors and fuse the context features and local features extracted by BiLSTM and IDCNN respectively. The proposed model is tested on Chinese Electronic Medical Record (EMR) dataset issued by China Conference on Knowledge Graph and Semantic Computing 2020 (CCKS2020).Compared with the baseline models such as BiLSTM-CRF, the experiment on CCKS2020 data shows that BERT-BiLSTM-IDCNN-Attention-CRF achieves 1.27% improvement in F1. The experimental results show that the proposed model can better identify the medical entities in EMR.

In order to obtain travel advice and route information more intuitively, efficiently and intelligently, it's necessary to establish a more intelligent information retrieval way. By defining events in the tourism field, this paper constructs the vocabulary of sequential cue phrases, completes the explicit sequential relationship recognition based on pattern matching, and stores the explicit sequential relationship in Neo4j graph database in the form of triples to establish an Eventic Graph. Based on constructed Eventic Graph, the tourism sequential Eventic Graph application system has been designed and established. With the layered B/S structure design, the system is based on Flask framework, and the query function of the tourism route based on city and planning days is realized. Using D3.js to visually display the Eventic Graph by the graphical method. The experimental results show that the accuracy rate of explicit sequential relationship extraction based on pattern matching reaches 82.84%. The visual query function based on Eventic Graph can help people to retrieve tourist information more efficiently and intuitively, and provide references for users to make decision.

针对传统字向量难以表达上下文语义以及抽取的特征较为单一等问题,提出基于BERT的多特征融合模型BERT-BiLSTM-IDCNN-Attention-CRF,通过BERT建模字向量的上下文语义关系,并融合双向长短期记忆网络(BiLSTM)和迭代膨胀卷积(IDCNN),分别抽取的上下文特征和局部特征,使两种特征进行互补以提升实体抽取效果.本模型在全国知识图谱与语义计算大会CCKS2020中文电子病历数据集上进行测试,与BiLSTM-CRF等基准模型进行比较,F1值提升1.27％.实验结果表明,本模型能较好地识别电子病历中的医疗实体.

中文新闻事件检测的主要任务是从大量新闻中自动检测出潜在的新事件.人工构建事件特征词进行检测费时费力.单纯依靠密度聚类或谱聚类方法进行事件检测,存在不同事件的触发词相关性高时,容易误判为同一事件.为此,提出基于卷积神经网络和K-means结合的中文新闻事件检测与主题提取,将新闻中的文本向量化,通过深度学习抽取文本深层特征.实验结果表明:所提方法构建的模型准确率、召回率优于单一聚类方法,可对中文新闻事件准确识别,快速检测新事件,提取新事件主题.

在新工科与"卓工"2.0背景下,针对网络工程专业面临内涵建设与人才培养转型升级的问题,提出地方高校通过产学合作促进网络工程专业综合改革,提升专业教育教学质量和创新人才培养机制,以青岛大学与网络企业开展产学合作协同育人为例,阐述地方高校通过教育部产学合作协同育人项目优化专业课程知识体系、丰富专业课程实验内容、提升专业教师实践技能培训、构建专业学生"阶梯式竞赛"梯队.

为了能够更快速、准确、智能地获取企业信息,需建立更智能的数据组织方式.通过构建企业领域本体,利用Karma建模实现多源异构数据的快速集成并发布成统一的RDF数据,并应用知识推理技术发现约20％的新知识,更新扩充RDF数据并存入Neo4j图数据库建立知识图谱.使用可视化技术设计实现企业知识图谱查询系统,用图形方法将知识图谱信息直观展现,以Web服务的方式提供企业信息检索服务,实现了对企业、法人及其相互关系的查询功能.实验结果表明,企业知识图谱明显提高了数据获取效率,增强了数据的可用性、可理解性与可见性.

Based on the integration of multi-source data, an approach of domain-specific knowledge graph construction is proposed to guide the construction of a “people-centered” poverty alleviation knowledge graph, and to achieve cross-functional and cross-regional sharing and integration of national basic data resources and public services. Focusing on “precise governance and benefit people service”, poverty alleviation ontology is constructed to solve semantic heterogeneity in multiple data sources integration, and provide an upper data schema for poverty alleviation knowledge graph construction. Karma modeling is used to implement semantic mapping between ontology concepts and data, and integrate multi-source heterogeneous data into RDF data. The RDF2Neo4j interpreter is developed to parse RDF data and store RDF data schema based on the graph database Neo4j. Based on visualization technology and natural language processing technology, Poverty Alleviation Knowledge Graph Application System is designed to achieve knowledge graph query and knowledge question answering function, which improved the application value of government data.

The purpose of data integration is that integrates multi-source heterogeneous data. Ontology solves semantic describing of multi-source heterogeneous data. The authors propose a practical approach based on ontology modeling and an information toolkit named Karma modeling for fast data integration, and demonstrate an application example in detail. Armed Conflict Location & Event Data Project (ACLED) is a publicly available conflict event dataset designed for disaggregated conflict analysis and crisis mapping. The authors analyzed the ACLED dataset and domain knowledge to build an Armed Conflict Event ontology, then constructed Karma models to integrate ACLED datasets and publish RDF data. Through SPARQL query to check the correctness of published RDF data. Authors design and developed an ACLED Query System based on Jena API, Canvas JS, and Baidu API, etc. technologies, which provides convenience for governments and researches to analyze regional conflict events and crisis early warning, and it verifies the validity of constructed ontology and the correctness of Karma modeling.

At present, the construction of knowledge graph for poverty alleviation is relatively scarce. This paper builds a set of tools for semi-automatic generation of knowledge graph based on the data sources in the poverty alleviation field in Chongqing which. First, the LOAD statement is automatically generated and imported into the Neo4j graph database. Secondly, semantic mapping is performed between different data classes in the Chongqing poverty alleviation data source according to the rules defined by the ontology. Based on the results obtained by the semantic mapping, the MATCH statement imported graph database is automatically generated. Finally, query the relational data stored in the Neo4j graph database through the Echarts component. Users and developers can use and maintain knowledge graph conveniently through a feature-rich application interface, which is of great significance to the application and disclosure of basic data for poverty alleviation departments.

At present,the construction of knowledge maps for poverty alleviation is relatively scarce. This paper is based on the data sources in the poverty alleviation field in Chongqing,and builds a set of tools for semi-automatic generation of knowledge maps. First,the LOAD statement is automatically generated and imported into the Neo4j graph database. Secondly,semantic map?ping is performed between different data classes in the Chongqing poverty alleviation data source according to the rules defined by the ontology. Based on the results obtained by the semantic mapping,the MATCH statement import graph database is automatically generated. Finally,the keyword is entered,through the Echarts component to achieve the map search function by character,rela?tionship query. Users and developers can use and maintain knowledge maps conveniently and transparently through a feature-rich application interface,which is of great significance to the application and disclosure of basic data for poverty alleviation depart?ments.

Character relationship information is important for precision poverty alleviation. The main challenge is that the character relation data is heterogeneous and relation words are implicit in text short sentences. Therefore, we propose a method of character relationship mining based on the combination of knowledge graph and deep learning. The DNN and BiGRU neural network joint method are proposed to recognize person named entities and extract character relations. The character relation triples data are stored in the Neo4j graph database, and the deeper information is mined by using the multi-depth relation query method. The experimental results on the poverty alleviation data show that the best accuracy of relation extraction can reach 84.9%, and the query efficiency of character relationship is improved. Users could quickly view the implied character relationships through the graph.

针对当前高校网络工程专业人才培养存在的问题,提出通过产学合作,调整优化网络工程专业主干课程知识体系,梳理重构课程教学内容,丰富课程实验和课程设计案例,以适应"万物互联"时代高级网络技术人才培养的需要;以计算机网络原理课程为例,阐述如何将思科网络技术学院课程教学内容和实验案例,融入课程教学内容、丰富实践教学案例、建设优质课程教学资源,最后说明教学实践效果.

The design and planning of small-scale campus network is an important experimental content of comprehensive skills training.This paper analyzes key technologies of campus network construction,and uses Cisco Packet Tracer software to simulate the campus network topology planning,network equipment selection,network equipment interconnection configuration command,and finally achieves the basic functions of campus network.The teaching practice proves that using Cisco Packet Tracer can improve students' interest in network experimental design.Comprehensive design case teaching enhances students' ability to analyze and solve problems,cultivates students' innovative thinking,and achieves good teaching effect in practice teaching.

Aiming at multi-source heterogeneous data integration, this paper proposed an approach based on ontology-modeling and Karma-modeling, and applied this approach to integrating food security data.By analyzing the characteristics of food security relative datasets and domain knowledge, we extracted main concepts and relations to build food security ontology.Authors describe Karma, an ontology-based integrated toolkit that implements this approach and show how Karma can be applied to integrating data in the food security domain.Based on published RDF data, authors developed Food Security Data Management System (FSDMS), which provided query, statistical analysis and comparison etc.functions.This RDF application system has positive effects on food security risk management;it also verifies the validity of food security ontology and Karma modeling.

An integration method based on ontology and Karma modeling is proposed for the multi-source heterogeneous characteristics of Web Service data, and it is applied to the field of weather information.Weather information ontology is constructed to describe semantic relations between multi-source heterogeneous data.Using an integration toolkit named Karma to invoke different weather service's APIs, we can get the corresponding weather data with the JSON format which might be dirty data.Karma converts JSON data into spreadsheets respectively and proposes a service model which users can refine interactively, which can implement the semantic mapping between ontology and different data sets.We build corresponding Karma models to integrate weather data and publish RDF data.Based on published RDF data, we validate the correctness of the ontology through SPARQL Querying and verify and Karma modeling.After integration, RDF data has a standardized syntax and semantics, which will facilitate the use and RE development of users and service integrators.

针对如何集成来源多样、数据异构、表示不同的大数据的问题,提出一种基于本体和Karma建模的大数据集成方案,将其应用于武装冲突事件数据的快速集成;分析武装冲突事件数据项目ACLED的数据集,构建武装冲突事件本体进行数据语义描述;使用数据集成开发工具Karma针对多源异构数据建立相应的Karma模型完成数据的语义映射、数据清洗和整理,实现多源异构数据的快速集成;编写SPARQL查询验证了集成发布的RDF数据和Karma建模的正确性,也进一步验证了基于本体和Karma建模实现多源异构数据集成方法的有效性.

计算机网络原理是计算机专业的核心必修课程,课程中包含许多抽象的网络协议、过程原理和算法.本文研究Authorware的多媒体开发环境和编程功能,设计开发一个简单实用、功能强大、交互性强的计算机网络原理多媒体课件.阐述了用Authorware制作多媒体课件的设计框架、设计思路和方法步骤,并以计算机网络原理的三种信息交换过程、停止等待协议和使用子网划分的分组转发这三个典型协议和过程为例,具体讨论了使用Authorware设计实现中的图标技巧.教学实践证明,通过多媒体课件模拟仿真实现网络协议和过程原理,可以将晦涩难懂的网络协议和过程原理直观化和形象化,增强了学生学习的交互性、自主性和创造性.

MOOC的发展为高等教育课程改革带来了一场革命，翻转课堂作为一种新的课堂教学模式冲击着传统的课堂教学方式。文章分析MOOC和翻转课堂带来的教学模式的转变，探讨如何将MOOC和翻转课堂教学模式的理念融入“计算机网络原理”精品课程的建设，梳理课程知识体系，进行多样化的课程设计，制作课程视频和微课，开发课程辅助教学系统，等等；提出以教师重点知识引导与学生自主学习相结合的教学模式，旨在培养学生创新能力以适应互联网时代大学教育转型改革的趋势。