基于Word2Vec和LDA主题模型的中国省级五年规划“文化政策”文本研究

运用Word2Vec和LDA相结合的主题模型分析技术,对我国31个省份三个时期五年规划文本中文化政策部分进行主题识别,从时间和空间两个维度进行“文化政策”主题挖掘和演化分析。研究发现,“文化政策”主题在发展趋势、重点转移、政策导向、技术应用等方面随时间推移呈现不同演化趋势;四大区域受经济发展水平、文化资源禀赋、政策导向影响,在企业角色强调程度、地区特色旅游发展以及国家级项目和竞争力方面存在地域差异。

Python实现Excel文档转换到Word文档的自动化方法

通过研究开发了一种基于Python语言,实现Excel数据自动转换成Word文档并实现排版功能的方法,为科研工作者、数据分析师或报告撰写人员提供一个方便快捷的工具,帮助他们更高效地完成Excel数据转换成Word并实现排版的任务。

Clay minerals and elemental composition of sediments on different sedimentary units in the northern East China Sea shelf:provenance tracing and genetic mechanism analysis

The composition,provenance,and genetic mechanism of sediment on different sedimentary units of the East China Sea(ECS)shelf are essential for understanding the depositional dynamics environment in the ECS.The sediments in the northern ECS shelf are distributed in a ring-shaped distribution centered on the southwestern Cheju Island Mud.From the inside to the outside,the grain size goes from fine to coarse.Aside from the“grain size effect”,hydrodynamic sorting and mineral composition are important restrictions on the content of rare earth elements(REEs).Based on the grain size,REEs,and clay mineral composition of 300 surface sediments,as well as the sedimentary genesis,the northern ECS shelf is divided into three geochemical zones:southwestern Cheju Island Mud Area(ZoneⅠ),Changjiang Shoal Sand Ridges(ZoneⅡ-1),Sand Ridges of the East China Sea shelf(ZoneⅡ-2).The northern ECS shelf is mostly impacted by Chinese mainland rivers(the Changjiang River and Huanghe River),and the provenance and transport mechanism of sediments of different grain sizes is diverse.The bulk sediments come primarily from the Changjiang River,with some material from the Huanghe River carried by the Yellow Sea Coastal Current and the North Jiangsu Coastal Current,and less from Korean rivers.Among them,surface sediments in the southwestern Cheju Island Mud Area(ZoneⅠ)come mostly from the Changjiang River and partly from the Huanghe River.It was formed by the counterclockwise rotating cold eddies in the northern ECS shelf,which caused the sedimentation and accumulation of the fine-grained sediments of the Changjiang River and the Huanghe River.The Changjiang Shoal Sand Ridges(ZoneⅡ-1)were developed during the early-middle Holocene sea-level highstand.It is the modern tidal sand ridge sediment formed by intense hydrodynamic action under the influence of the Yellow Sea Coastal Current,North Jiangsu Coastal Current,and Changjiang Diluted Water.The surface sediments mainly originate from the Changjiang River and Huanghe River,with the Changjiang River dominating,and the Korean River(Hanjiang River)influencing just a few stations.Sand Ridges of the East China Sea shelf(ZoneⅡ-2)are the relict sediments of the paleo-Changjiang River created by sea invasion at the end of the Last Deglaciation in the Epipleistocene.The clay mineral composition of the surface sediments in the study area is just dominated by the Changjiang River,with the North Jiangsu Coastal Current and the Changjiang Diluted Water as the main transporting currents.

基于LDA和Word2Vec模型的学位论文评阅意见主题挖掘与分析

选取某高校部分硕士学位论文评阅意见为研究对象,使用自然语言处理和机器学习技术进行自动化的硕士学位论文评阅意见主题挖掘与分析。首先,采用LDA(latent dirichlet allocation)模型对评阅数据进行主题建模,提取文本中的潜在主题,并将评阅意见转化为主题分布向量;其次,结合Word2Vec模型将评阅意见的关键词转化为向量表达;最后,采用TextRank方法提取关键词,以揭示评阅专家的关注核心主题。实验结果表明,所提方法能为高校管理人员提供切实有效的分析工具,有助于他们更好地分析总结评阅意见,同时也为硕士研究生撰写高质量学位论文提供有益借鉴。

基于LDA-Word2vec的图书情报领域机器学习研究主题演化与热点主题识别

[目的/意义]在人工智能技术及应用快速发展与深刻变革背景下,机器学习领域不断出现新的研究主题和方法,深度学习和强化学习技术持续发展。因此,有必要探索不同领域机器学习研究主题演化过程,并识别出热点与新兴主题。[方法/过程]本文以图书情报领域中2011—2022年Web of Science数据库中的机器学习研究论文为例,融合LDA和Word2vec方法进行主题建模和主题演化分析,引入主题强度、主题影响力、主题关注度与主题新颖性指标识别热点主题与新兴热点主题。[结果/结论]研究结果表明,(1)Word2vec语义处理能力与LDA主题演化能力的结合能够更加准确地识别研究主题,直观展示研究主题的分阶段演化规律;(2)图书情报领域的机器学习研究主题主要分为自然语言处理与文本分析、数据挖掘与分析、信息与知识服务三大类范畴。各类主题之间的关联性较强,且具有主题关联演化特征;(3)设计的主题强度、主题影响力和主题关注度指标及综合指标能够较好地识别出2011—2014年、2015—2018年和2019—2022年3个不同周期阶段的热点主题。

基于Word VBA辅助技术文件编制的数字化协同建设的探索

文件编制是设计研发人员日常工作的重要组成部分,其中技术文件的编制涉及大量文件结构和起草规则的应用,基于相关标准中关于文件的编写要求,文章利用WordVBA编程技术辅助技术文件编写工作中的格式编排,实现了Word文档标准格式技术文件的自动化编制,从而保障文件编制符合标准格式要求,有效提高了工作效率。

结合Word2vec和BiLSTM的民航非计划事件分析方法

安全是民航业的核心主题。针对目前民航非计划事件分析严重依赖专家经验及分析效率低下的问题,文章提出一种结合Word2vec和双向长短期记忆(bidirectional long short-term memory,BiLSTM)神经网络模型的民航非计划事件分析方法。首先采用Word2vec模型针对事件文本语料进行词向量训练,缩小空间向量维度;然后通过BiLSTM模型自动提取特征,获取事件文本的完整序列信息和上下文特征向量;最后采用softmax函数对民航非计划事件进行分类。实验结果表明,所提出的方法分类效果更好,能达到更优的准确率和F 1值,对不平衡数据样本同样具有较稳定的分类性能,证明了该方法在民航非计划事件分析上的适用性和有效性。

Provenance and Paleoclimate of the Triassic to Middle Jurassic Adigrat Sandstone,Blue Nile Basin,Central Ethiopia

Mineralogical and geochemical studies have been undertaken on the Triassic to Lower Jurassic Adigrat Sandstone of the Blue Nile Basin of central Ethiopia to infer its source rock type,paleoweathering,and paleoclimatic history.The Adigrat Sandstone occurs at the basal section of the Mesozoic sedimentary formation and unconformably overlays the Neoproterozoic–Paleozoic crystalline rocks,or locally,the Karroo sediments in the northern Blue Nile Basin.A mineralogical study reveals that quartz(Q),feldspars(F),and lithic fragments(L)are the framework grains of the sandstone.On the QFL diagram,the plot of the modal composition of the sandstone mainly falls within the feldspathic arenite and quartzose arenite fields.The geochemical data of the lower section of the sandstone mainly falls within the arkose and subarkose fields,whereas the upper section data falls within the quartzose and sublithic arenite fields.Mineralogical and geochemical weathering indices indicate that the provenances of the Adigrat Sandstone were exposed to pronounced weathering intensity,where the lower part of the sandstone was controlled by arid to semi-arid conditions,whereas the upper section was linked to humid to semi-humid(tropical to subtropical)climatic conditions.Mineralogical and geochemical data also indicate that mafic to intermediate basement rocks were the primary source rocks of the sediment.Perhaps the sediment was assumed to have been reworked by multi-cyclic sedimentary processes.The discriminant function diagram,the REE pattern,La/Th vs.La/Yb,and the Th–Hf–Co plot are consistent.A comparison of provenance studies for the Adigrat Sandstone in the Blue Nile Basin and the Mekele outlier of northern Ethiopia indicates that the sediment of the former is highly sorted,experienced higher weathering intensity,and compositionally derived from mafic to intermediate crystalline rocks.On the other hand,the sediment of the latter is essentially a weathering product of felsic rocks.

基于Word2Vec和决策树的故障定位技术

利用Word2Vec方法对Java源代码进行深层语义编码,生成文件级和行级的语义向量,并将其用作输入数据来训练决策树模型,以实现精确的文件级别和行级别故障定位,优化故障检测过程,构建一个综合文件级别与行级别分析的高效故障定位框架.实验结果表明:该模型在各项目中的故障定位准确率均高于83%.

关于Word2Vec文本分类效果若干影响因素的分析

Word2Vec向量模型参数众多,在不同情景下分类效果不一,分析其影响因素很有必要。从Word2Vec模型基本原理出发,分析讨论了预训练语料、词向量预训练参数以及分类模型参数三大因素对模型分类效果的影响。结果表明限定域预料效果好于广域预料;预训练参数中向量维度越大,效果越好,窗口大小存在最优值,分类算法影响不大;分类模型参数中学习率、激活函数、批次大小对模型分类效果影响较大,训练轮次相对较小。

Provenance of Conglomerate and Sandstone from Early Permian Shoushangou Formation in Xi Ujimqin,Inner Mongolia:Implications for Understanding Paleo-Asian Ocean Subduction

During the Late Carboniferous to Early Permian,a rift was formed by post-collisional extension after ocean closure or an island arc-related basin formed by Paleo-Asian Ocean(PAO)subduction in the Xi Ujimqin area.Nevertheless,the closure time of the PAO is still under debate.Thus,to identify the origin of the PAO,the geochemistry and U-Pb age of zircons were analyzed for the extra-large deep marine,polymict clastic boulders and sandstones in the Shoushangou Formation within the basin.The analyses revealed magmatic activity and tectonic evolution.The conglomerates include megaclasts of granite(298.8±9.1 Ma)and granodiorite porphyry(297.1±3.1 Ma),which were deposited by muddy debris flow.Results of this study demonstrated that the boulders of granitoids have the geochemistry of typical I-type granite,characterized by low Zr+Nb+Ce+Y and low Ga/Al values.The granitoid boulders were formed in island arc setting,indicating the presence of arc magmatism in the area that is composed of the Late Carboniferous to Early Permian subduction-related granitoid in southern Xi Ujimqin.Multiple diagrams for determining sedimentary provenance using major and trace elements indicate that Shoushangou sediments originated from continental island arc-related felsic rocks.Detrital zircon U-Pb age cluster of 330–280 Ma was obtained,indicating input from granite,ophiolite,Xilin Gol complex,and Carboniferous sources to the south.The basin was geographically developed behind the arc during the Early Permian period because the outcropped intrusive rocks in the Late Carboniferous to Early Permian form a volcanic arc.The comprehensive analyses of source areas suggest that Shoushangou sediments developed in a backarc basin in response to the northward subduction of the PAO.The backarc basin and intrusive rocks,in addition to previously published Late Carboniferous to Early Permian magmatic rocks of arc unit in Xilin Gol,confirm the presence of an Early Permian trencharc-basin system in the region,represented by the Baolidao arc and Xi Ujimqin backarc basin.This study highlights the importance and potential of combined geochemical and geochronological studies of conglomerates and sandstone for reconstructing the geodynamic setting of a basin.

标准电子文件转WORD校验方法

随着标准电子文件在各个领域的广泛应用,文件格式的转换及其后续校验工作变得日益重要。本研究聚焦于探讨标准文档格式转换为Word格式的过程中所面临的挑战,并提出了一个全面的校验流程,以确保转换的准确性和高效性。阐述了标准电子文件在石油等领域的广泛应用以及转换为Word格式的必要性,强调了校验过程的重要性,旨在确保转换后的文件与原始文件在错误率万分之五以内的一致性。

基于K-means与Word2vec的哺乳文胸评论主题挖掘研究

目的为了了解消费者在网络平台购买哺乳文胸时的关注侧重点,文章从在线评论中抽取有效关键词构建哺乳文胸主题,并通过计算主题的重要程度协助商家了解消费者关注重点方向。方法选用TF-IDF关键词抽取算法,结合K-means和Word2vec进行语义聚类、主题识别、主题词挖掘及主题重要度计算。结果哺乳文胸评论文本聚类后的主题重要程度排名是:产品品质(45.47%)、产品外观(35.83%)、产品服务(18.79%)。结论通过该方法能够有效的识别和构建哺乳文胸主题及主题词,同时,通过主题的重要程度,能够了解消费者对于网络平台购买哺乳文胸时关注的重点方向,为哺乳内衣企业进行产品改善及生产等提供理论参考。

Sentiment Analysis Using E-Commerce Review Keyword-Generated Image with a Hybrid Machine Learning-Based Model

In the context of the accelerated pace of daily life and the development of e-commerce,online shopping is a mainstreamway for consumers to access products and services.To understand their emotional expressions in facing different shopping experience scenarios,this paper presents a sentiment analysis method that combines the ecommerce reviewkeyword-generated imagewith a hybrid machine learning-basedmodel,inwhich theWord2Vec-TextRank is used to extract keywords that act as the inputs for generating the related images by generative Artificial Intelligence(AI).Subsequently,a hybrid Convolutional Neural Network and Support Vector Machine(CNNSVM)model is applied for sentiment classification of those keyword-generated images.For method validation,the data randomly comprised of 5000 reviews from Amazon have been analyzed.With superior keyword extraction capability,the proposedmethod achieves impressive results on sentiment classification with a remarkable accuracy of up to 97.13%.Such performance demonstrates its advantages by using the text-to-image approach,providing a unique perspective for sentiment analysis in the e-commerce review data compared to the existing works.Thus,the proposed method enhances the reliability and insights of customer feedback surveys,which would also establish a novel direction in similar cases,such as social media monitoring and market trend research.

浅谈Word软件在编辑工作中的应用——以Word2007版本为例

Word软件具有功能强大、操作便捷等特点,是应用最为广泛的文字处理软件之一。文章以Word2007版本为例,介绍编辑工作中常用的Word功能,希望对从事编辑工作的人员和其他想了解Word软件的人员有所帮助。

Vice-President Ai Ping Meets the US Word4Asia Delegation in Beijing

On March 21st,Mr.Ai Ping,Vice-President of CAFIU,met in Bejing with a 6-member delegation from the US Word4Asia led by Dr.Gene Wood,Chief Executive Officer of Word4Asia.Mme.Xiao Ningning,Deputy Secretary-General of China NGO Network for International Exchanges,Mr.Gu Chuanyong,Vice-Chair of Board of the Amity Foundation,and Mme.She Hongyu,Deputy Secretary-General of the Amity Foundation,attended the meeting.The two sides exchanged views on deepening people-to-people exchanges and cooperation between China and the United States and on the preparation of the 5th"Amity Cup"International Table Tennis Charity Competition.

Word试题自动阅卷研究

关于通过VBA进行Word自动阅卷的论文多是介绍很基本的内容,所举的例子十分简单。其实VBA本是为宏推出的,用于自动阅卷,有些题目并不易实现,有必要进行探讨。文章举几个具体的例子。

利用Word VBA辅助书稿编校的研究

VBA是一种面向对象的、可视化的、事件驱动的编程语言,是Windows应用程序通用的宏控制语言,语法较为简单,但它能够在Word、Excel等办公软件中高效地处理一些具有显著特征的、程序性工作。在书稿编校中,不少工作属于具有一定特征的重复性简单劳动,如文字段落排版、词语查找替换、图表号检查等。如果稿件篇幅较长,仅靠人工逐个处理,工作量大,也非常耗时,而且容易遗漏甚至出现错误。如果此类工作交由Word VBA进行辅助批量处理,则可大幅提高工作效率,有效降低编辑工作错误率。文章首先对Word VBA作简要概述,列出VBA在Office办公软件中的一些用途,并介绍使用方法;接着对VBA中非常重要的、使用较多的正则表达式作简要概述,列举了一些正则表达式匹配实例,并介绍了用于有效测试正则表达式准确性的正则表达式测试工具,以帮助读者更好地编写和测试复杂的正则表达式;然后以文字段落排版、批量替换用词、集中汇总检查序号等为例,具体阐述Word VBA在书稿辅助编校中的应用,同时给出相应程序的核心代码,以供读者参考。文章最后指出,Word VBA在书稿编校中应用广泛,工作中应根据书稿特点编写出通用性较强的程序,并在使用中测试完善,以不断提高书稿编校工作的效率。

计算机办公软件Word的具体操作应用探析

在办公软件中,Word作为一种基础软件,主要用于文字编辑、排版等方面,但是其功能多样,如果操作不熟练则会影响到软件功能的正常使用。鉴于此,文章主要围绕计算机办公软件Word的实际操作应用内容展开介绍,以期能够为相关人员的工作开展提供借鉴和参考。

A Contrastive Analysis of English and Chinese Idioms About Color Words & Their Cultural Connotations

There are many idioms related to color words in English and Chinese.The use of color words in idioms adds beauty and vividness to the language.Due to the cultural differences,“color idioms”have gained different cultural connotations with the development of English and Chinese languages.It is of great significance to accurately understand and grasp the meanings and differences of color-related idioms in Chinese and English.This paper intends to analyze and expound the cultural connotations of English and Chinese idioms related to several widely used basic color words with the aim of helping English learners know and use the idioms about color words better.