Identification and distribution of 13003 landslides in the northwest margin of Qinghai-Tibet Plateau based on human-computer interaction remote sensing interpretation

The periphery of the Qinghai-Tibet Plateau is renowned for its susceptibility to landslides.However,the northwestern margin of this region,characterised by limited human activities and challenging transportation,remains insufficiently explored concerning landslide occurrence and dispersion.With the planning and construction of the Xinjiang-Tibet Railway,a comprehensive investigation into disastrous landslides in this area is essential for effective disaster preparedness and mitigation strategies.By using the human-computer interaction interpretation approach,the authors established a landslide database encompassing 13003 landslides,collectively spanning an area of 3351.24 km^(2)(36°N-40°N,73°E-78°E).The database incorporates diverse topographical and environmental parameters,including regional elevation,slope angle,slope aspect,distance to faults,distance to roads,distance to rivers,annual precipitation,and stratum.The statistical characteristics of number and area of landslides,landslide number density(LND),and landslide area percentage(LAP)are analyzed.The authors found that a predominant concentration of landslide origins within high slope angle regions,with the highest incidence observed in intervals characterised by average slopes of 20°to 30°,maximum slope angle above 80°,along with orientations towards the north(N),northeast(NE),and southwest(SW).Additionally,elevations above 4.5 km,distance to rivers below 1 km,rainfall between 20-30 mm and 30-40 mm emerge as particularly susceptible to landslide development.The study area’s geological composition primarily comprises Mesozoic and Upper Paleozoic outcrops.Both fault and human engineering activities have different degrees of influence on landslide development.Furthermore,the significance of the landslide database,the relationship between landslide distribution and environmental factors,and the geometric and morphological characteristics of landslides are discussed.The landslide H/L ratios in the study area are mainly concentrated between 0.4 and 0.64.It means the landslides mobility in the region is relatively low,and the authors speculate that landslides in this region more possibly triggered by earthquakes or located in meizoseismal area.

Research on the Man-machine Interactive Environment VR and the Applications on Vocational Education and Training under the Perspective of Interactivity

In this paper, we conduct research on the man-machine interactive environment VR and the applications on vocational educationand training under the perspective of interactivity. With the increase in the general standard of social knowledge level and competition intensifi es,more and more people have a goal to build a lifelong learning system, according to their own hobbies, work and the needs of the marketcompetition. Under this condition, the vocational education is becoming more and more essential. This paper integrates the VR and man-machineinteractive concept to propose the new education paradigm that is innovative.

Research on the Interactive Interpreting Teaching Mode under the Multimedia and Traditional Microteaching Environment

In this paper, we conduct research on the interactive interpreting teaching mode under the multimedia and primary traditional microteaching environment. English interpretation course system construction request from the student reality, on the basis of the actual situation, improving students＇ comprehensive ability in English interpretation as communicative teaching method in the very great degree to overall effective guide students to practice real interpreting scene comprehensively improve students＇ intercultural business negotiation, reception and comprehensive ability of the exhibition, and is advantageous to the interpretation of curriculum system construction and the improvement of the students＇ comprehensive quality. Our research integrates the general principles of the multimedia and primary traditional microteaching environment to form the better paradigm for the education that will enhance the contemporary teaching condition.

Research on menu-typed man-machine interaction system of digital TIG welding machine

A digital man-machine interaction system controlled by communications between two processors of TMS320F240 and AT98C2051 was researched in the paper. The system is easy to set and modify welding process parameters by keyboards, and display information of welding site by LCD (Liquid Crystal Display). As one part of multi-task system about TIG welding machine, the coordination of man-machine interaction system with other tasks is the main point to the stability and reliability of its operation. Experiments result indicates that the system is stable, operation-flexible, high precision, and anti-interfering.

强化学习的可解释方法分类研究

强化学习能够在动态复杂环境中实现自主学习,这使其在法律、医学、金融等领域有着广泛应用。但强化学习仍面临着全局状态空间不可观测、对奖励函数强依赖和因果关系不确定等诸多问题,导致其可解释性弱,严重影响其在相关领域的推广,会遭遇诸如难以判断决策是否违反社会法律道德的要求,是否准确及值得信任等的限制。为了进一步了解强化学习可解释性研究现状,从可解释模型、可解释策略、环境交互、可视化等方面展开讨论。基于此,对强化学习可解释性研究现状进行系统论述,对其可解释方法进行归类阐述,最后提出强化学习可解释性的未来发展方向。

用者为本,技之大道--评《以口译员为中心的技术可用性研究:以SmarTerp为例》

在口译学科“技术转向”的时代背景下,口译技术人性化和可持续发展的重要性日益凸显。《以口译员为中心的技术可用性研究:以SmarTerp为例》一书立足时代需求,介绍了全球首项使用可用性工程方法开发计算机辅助口译工具的案例研究,旨在反映口译领域技术变革,确保口译技术的公平可持续发展。该书通过采用人机交互的理论框架和方法论框架评估计算机辅助口译工具SmarTerp界面的可用性,弥补了当前计算机辅助口译实证研究的不足,对口译技术的研发和应用具有重要的启发意义和借鉴价值。

面向心律失常疾病的临床决策支持系统交互性研究与实现

目的/意义针对当前临床决策支持系统使用过程中存在的问题,提出界面设计原则并将其应用于心律失常疾病临床决策支持系统建设中。方法/过程从交互设计角度,针对可解释性、时效性、可用性、相关性、尊重性和循证性6个维度,提出界面设计原则。选取心律失常疾病决策支持作为临床场景,设计临床决策支持系统交互界面原型,阐述界面功能、信息功能、交互功能设计与实现过程。结果/结论本研究提出的界面设计原则可有效缓解6类问题,经论证具有应用于心律失常疾病临床决策支持原型系统的可行性以及可泛化性,可用于指导多种疾病辅助诊疗工具的交互设计。

面向任务的载人深潜人机交互信息重要度评估

为更好地识别复杂任务环境下深海载人潜水器驾驶舱人机交互信息重要度差异,提出一种基于复杂网络的人机交互信息重要度评估方法。以悬停采样任务为研究对象,运用决策阶梯模型及任务-网络建模技术识别复杂任务相关人机交互信息元及逻辑关联,构建交互信息复杂网络模型。基于网络拓扑特征参数,从节点间逻辑影响关系的角度提出节点全局及局部影响效应评估指标,结合解释结构模型层级权重指标对交互信息节点重要度综合值进行计算,得出悬停采样任务人机交互信息重要度排序。将排序结果与其他4种算法进行对比,在证明计算有效的基础上,邀请潜航员结合实际任务流程对交互信息进行主观重要度评估。结果显示,研究所得重要度排序结果与潜航员主观排序结果呈强相关,且对重要交互信息元的识别结果与潜航员主观感知相符合,验证了所提方法识别复杂任务人机交互信息重要度差异的有效性。

社会主义核心价值观融入法学专业课程的价值、作用与路径

社会主义核心价值观融入法学专业课程,既是新时代培养德法兼修的卓越法治人才的内在要求,也是深入推进法治中国建设的外在需求。社会主义核心价值观继承了中华优秀传统文化,践行了习近平法治思想,为培养新时代德法兼修的卓越法治人才奠定了优良的价值基础。社会主义核心价值观与法学专业课程的教学目标高度契合,法学教师可以从理论、实践双层面实现社会主义核心价值观浸润式融入法学专业课程,并通过多维度评价机制实现对学生的考核反馈,形成社会主义核心价值观与法学专业课程的相互作用,培养出社会主义核心价值观的合格践行者。

人智交互情境下的知识转化模型:内涵深化及外延拓展

[目的/意义]交互式人工智能时代下,机器对信息的处理达到人类难以触及的新高度,暗知识的产生对知识管理提出了新的要求,知识转化模型(SECI Model)已经难以诠释交互式人工智能时代知识管理的内涵及外延,亟须拓展更新以适应技术发展的新要求。[方法/过程]首先回顾SECI模型以及场(Ba)理论的发展沿革,随后结合人智交互特征讨论了知识管理的内涵与外延以及知识转化过程的拓展与变革,对知识管理模型进行进一步的思考与总结,并针对暗知识与显性知识相互转化的过程提出发展进路。[结果/结论]在SECI模型的基础上提出了人智交互情境下新的知识转化模型(T-E-D Model)。提出显性知识是人智交互中暗知识与隐性知识链接的桥梁,暗知识与显性知识的转化是知识管理新情境下的关键,需要从基础设施、过程管理、个性化设计以及标准制定4个方面推进暗知识转化的质量与效率。[局限]所提出理论模型仅建立在交互式人工智能作为认知主体的语境下,而AI能否作为独立于人脑外的认知主体还需进一步的伦理探索。

Interpretability of Neural Networks Based on Game-theoretic Interactions

This paper introduces the system of game-theoretic interactions,which connects both the explanation of knowledge encoded in a deep neural networks(DNN)and the explanation of the representation power of a DNN.In this system,we define two gametheoretic interaction indexes,namely the multi-order interaction and the multivariate interaction.More crucially,we use these interaction indexes to explain feature representations encoded in a DNN from the following four aspects:(1)Quantifying knowledge concepts encoded by a DNN;(2)Exploring how a DNN encodes visual concepts,and extracting prototypical concepts encoded in the DNN;(3)Learning optimal baseline values for the Shapley value,and providing a unified perspective to compare fourteen different attribution methods;(4)Theoretically explaining the representation bottleneck of DNNs.Furthermore,we prove the relationship between the interaction encoded in a DNN and the representation power of a DNN(e.g.,generalization power,adversarial transferability,and adversarial robustness).In this way,game-theoretic interactions successfully bridge the gap between“the explanation of knowledge concepts encoded in a DNN”and"the explanation of the representation capacity of a DNN"as a unified explanation.

基于语用交互的跨目标立场检测

针对缺乏足够的带标注意见数据、跨目标立场检测结果不佳且可解释性弱等问题,提出一种基于语用交互(pragmatic interaction graph convolution, PIGCN)的跨目标立场检测模型。考虑情感与立场在语义上的耦合关系,利用交互式图卷积神经网络(graphical convolutional network, GCN),增量式聚合单词在不同目标之间语用信息的相互作用,缓解目标间的信息孤岛问题。实验结果表明,该模型在平均F1值上达到了53.4%,优于基准模型,具有更好的可扩展性和适应性,在提升模型可解释性方面具有潜力。

本科英语专业口译教学中的同伴互动合作学习研究

近年来,同伴互动合作学习在二语教学中受到了更多的关注,但在口译教学中的应用研究较少。口译作为一种语言文化的交流活动,其互动性在教学过程中尤其重要。本研究采用问卷调查和访谈法,检验本科英语专业口译课程中同伴互动合作学习的教学效果。经过一学期的教学实践,学生对该教学法接受程度高,同伴互动合作学习适用于口译课堂教学。

The Inspiration of The Interpreter's Roles to College English Teaching

Jef Verschueren is the representative of European Continental pragmatic thought.He puts forward a new theory that pragmatics is a perspective on language and he uses the theory of adaptation to analyze pragmatics."The general concern for the study of linguistic pragmatics is to understand the meaningful functioning of a language as a dynamic process operating on context-structure relationships at various levels of salience"(Verschuren 2000:69).The focal points in contexts are the utterer and the interpreter,for the physical,social and mental aspects of contexts are activated by language users’cognitive processes.Verschueren divided interpreter’s roles into five types:addressee,side participant,bystander,listener-in and eavesdropper,which can explain the adaptation of context and language choices.This is also applicable in college English classroom as the English class is also a context.At present students taught at the same time by the same teacher do not master the knowledge at the same level.One of the reason is that students are not engaged in the class at the same level.According to the Verschueren’s theory,the best English teaching class is one that each student in the addressee of teachers.Addressee is a metaphor which means everyone should follow the teachers’explanation carefully and take part in the teaching activities.It has something to do with interaction theory,which considers students as active individuals but not passive receivers.One of the common interaction patterns is questioning.In this thesis,the question how to make students addresses through adoption of questioning approach in college English class will be researched.

感知解说质量对游客博物馆体验的影响研究:基于互动仪式链与AR技术应用视角

随着数字技术应用于旅游领域的实践探索日益多元化,博物馆中传统解说服务难以满足游客的体验需求,依靠人机互动技术拉近人与物的距离对文化传承有重要的意义。增强现实技术(AR)以三维立体、虚实结合、实时交互的效果“复活”了博物馆展品,是对人工解说内容的具象化,既让游客深入了解展品,又能引发游客展开想象。但这种心理意象是否在游客感知解说质量到博物馆体验的路径中发挥作用、AR技术是否可以起到辅助人工解说的体验效果,这两个问题值得深入验证。本研究基于互动仪式链理论,以心理意象为中介变量,通过2组实验、3个情境对感知解说质量、心理意象、博物馆体验、AR技术应用4个核心要素的影响过程与边界条件进行了分析。研究发现:(1)3个情境下,心理意象在感知解说质量对游客博物馆体验的影响路径中均发挥中介作用;(2)人工解说叠加AR技术应用后,感知解说质量、心理意象、博物馆体验的作用路径更加清晰,有效提升了游客博物馆体验模型的解释度;(3)当人工解说发挥不稳定时,AR技术应用可以有效降低其造成的影响,进一步提升游客的博物馆体验。研究结论验证了互动仪式链和AR技术在博物馆情境中的积极作用,还原了心理意象在感知解说质量影响游客博物馆体验过程中的中介效应,丰富了游客博物馆体验模型,为博物馆高质量创新发展提供了理论依据。

浙学与实学的互诠互动

浙学与实学作为中华传统学术的重要组成部分,相生相伴、互诠互动。浙学的发展演变始终以实学形态呈现,一部浙学史就是传统实学发展历程的一个缩影。浙学的本质是实学,实学的理论特质在浙学的发展历程中也得以充分诠释。“实事疾妄”“崇实黜虚”“明体达用”“知行合一”“开物成务”“经世致用”,则是实学与浙学共有共享的优良学统与基本精神。

融合门控单元与多头自注意力机制的特征自动交互推荐算法

为了解决推荐算法中使用手工制作、特征工程等方式枚举所有的特征组合不但会带来巨大的存储空间和计算成本,而且无用的特征交互会引入噪声使模型训练过程复杂化的问题,文中提出融合多头自注意力机制的特征自动交互推荐算法。该算法首先利用门控机制对输入特征进行初次筛选;然后将特征送入多头自注意力机制中,选取关键特征进行不同阶的组合;最后利用残差网络进行特征融合输出预测结果。该算法能有效地提高预测结果的准确性,同时具有良好的解释性。

遥感技术在广东龙川县新田镇地质灾害调查的应用

地质灾害严重威胁社会经济建设和人们生命财产安全。为制定科学的防灾减灾策略,保障人民群众生命与财产安全。利用遥感技术宏观且全面的角度,通过多源遥感影像数据分析,结合区内已有地质灾害点和野外勘探,采用人机交互的综合方法,对广东龙川县新田镇进行岩性、构造、地质灾害解译,并分析地质灾害与发生时间、地形地貌、构造、人类活动的关系,确定滑坡、地质灾害隐患、崩塌等地质灾害的类型、规模及空间分布特征,结果表明,共解译地质灾害地灾点136处;地质环境点3处,且规模以小型为主;在时间上,集中在5—8月的雨季且具有滞后性;地质灾害整体发育程度北部大于南部,集中在380~460 m高程坡度低于30°坡向东北到北部的低山丘陵区;人类活动是地质灾害隐患点发育的主要因素。通过遥感地质灾害解译及实地验证,可有效识别和监测各类地质灾害情况,对新田镇地质灾害防治规划有指导意义。

基于深度学习的视觉问答研究综述

视觉问答是计算机视觉和自然语言处理的交叉领域。在视觉问答的任务中,机器首先需要对图像、文本这两种模态数据进行编码,进而学习这两种模态之间的映射,实现图像特征和文本特征的融合,最后给出答案。视觉问答任务考验模型对图像的理解能力以及对答案的推理能力。视觉问答是实现跨模态人机交互的重要途径,具有广阔的应用前景。最近相继涌现出了众多新兴技术,如基于场景推理的方法、基于对比学习的方法和基于三维点云的方法。但是,视觉问答模型普遍存在推理能力不足、缺乏可解释性等问题,值得进一步地探索与研究。文中对视觉问答领域的相关研究和新颖方法进行了深入的调研和总结。首先介绍了视觉问答的背景;其次分析了视觉问答的研究现状并对相关算法的和数据集进行了归纳总结;最后根据当前模型存在的问题对视觉问答的未来研究方向进行了展望。

基于Python的断面解释图目标层埋深数据提取程序开发

在电法勘探项目中,目标层埋深图是主要提交成果之一。这里结合反演电阻率推断解释断面图中常见的地质情况,分析总结了不同地质情况下的目标层埋深数据提取方式,并提出了圈定埋深数据边界的算法。利用Python语言及其第三方库的强大计算和绘图功能,实现了埋深数据的提取及其边界的圈定,为下一步数据成图提供借鉴。实践结果表明,该程序能有效地结合不同地质情况,借助图形进行交互操作,能快速、准确地提取目标层埋深数据,具有较强的实用性。