Mathematical Expression Extraction in Text Fields of Documents Based on HMM

Aiming at the problem that the mathematical expressions in unstructured text fields of documents are hard to be extracted automatically, rapidly and effectively, a method based on Hidden Markov Model (HMM) is proposed. Firstly, this method trained the HMM model through employing the symbol combination features of mathematical expressions. Then, some preprocessing works such as removing labels and filtering words were carried out. Finally, the preprocessed text was converted into an observation sequence as the input of the HMM model to determine which is the mathematical expression and extracts it. The experimental results show that the proposed method can effectively extract the mathematical expressions from the text fields of documents, and also has the relatively high accuracy rate and recall rate.

Semantic Document Layout Analysis of Handwritten Manuscripts

A document layout can be more informative than merely a document’s visual and structural appearance.Thus,document layout analysis(DLA)is considered a necessary prerequisite for advanced processing and detailed document image analysis to be further used in several applications and different objectives.This research extends the traditional approaches of DLA and introduces the concept of semantic document layout analysis(SDLA)by proposing a novel framework for semantic layout analysis and characterization of handwritten manuscripts.The proposed SDLA approach enables the derivation of implicit information and semantic characteristics,which can be effectively utilized in dozens of practical applications for various purposes,in a way bridging the semantic gap and providingmore understandable high-level document image analysis and more invariant characterization via absolute and relative labeling.This approach is validated and evaluated on a large dataset ofArabic handwrittenmanuscripts comprising complex layouts.The experimental work shows promising results in terms of accurate and effective semantic characteristic-based clustering and retrieval of handwritten manuscripts.It also indicates the expected efficacy of using the capabilities of the proposed approach in automating and facilitating many functional,reallife tasks such as effort estimation and pricing of transcription or typing of such complex manuscripts.

Environmental complaint insights through text mining based on the driver,pressure,state,impact,and response(DPSIR)framework:Evidence from an Italian environmental agency

Individuals,local communities,environmental associations,private organizations,and public representatives and bodies may all be aggrieved by environmental problems concerning poor air quality,illegal waste disposal,water contamination,and general pollution.Environmental complaints represent the expressions of dissatisfaction with these issues.As the timeconsuming of managing a large number of complaints,text mining may be useful for automatically extracting information on stakeholder priorities and concerns.The paper used text mining and semantic network analysis to crawl relevant keywords about environmental complaints from two online complaint submission systems:online claim submission system of Regional Agency for Prevention,Environment and Energy(Arpae)(“Contact Arpae”);and Arpae's internal platform for environmental pollution(“Environmental incident reporting portal”)in the Emilia-Romagna Region,Italy.We evaluated the total of 2477 records and classified this information based on the claim topic(air pollution,water pollution,noise pollution,waste,odor,soil,weather-climate,sea-coast,and electromagnetic radiation)and geographical distribution.Then,this paper used natural language processing to extract keywords from the dataset,and classified keywords ranking higher in Term Frequency-Inverse Document Frequency(TF-IDF)based on the driver,pressure,state,impact,and response(DPSIR)framework.This study provided a systemic approach to understanding the interaction between people and environment in different geographical contexts and builds sustainable and healthy communities.The results showed that most complaints are from the public and associated with air pollution and odor.Factories(particularly foundries and ceramic industries)and farms are identified as the drivers of environmental issues.Citizen believed that environmental issues mainly affect human well-being.Moreover,the keywords of“odor”,“report”,“request”,“presence”,“municipality”,and“hours”were the most influential and meaningful concepts,as demonstrated by their high degree and betweenness centrality values.Keywords connecting odor(classified as impacts)and air pollution(classified as state)were the most important(such as“odor-burnt plastic”and“odor-acrid”).Complainants perceived odor annoyance as a primary environmental concern,possibly related to two main drivers:“odor-factory”and“odorsfarms”.The proposed approach has several theoretical and practical implications:text mining may quickly and efficiently address citizen needs,providing the basis toward automating(even partially)the complaint process;and the DPSIR framework might support the planning and organization of information and the identification of stakeholder concerns and priorities,as well as metrics and indicators for their assessment.Therefore,integration of the DPSIR framework with the text mining of environmental complaints might generate a comprehensive environmental knowledge base as a prerequisite for a wider exploitation of analysis to support decision-making processes and environmental management activities.

Document Clustering Using Semantic Cliques Aggregation

The search engines are indispensable tools to find information amidst massive web pages and documents. A good search engine needs to retrieve information not only in a shorter time, but also relevant to the users’ queries. Most search engines provide short time retrieval to user queries;however, they provide a little guarantee of precision even to the highly detailed users’ queries. In such cases, documents clustering centered on the subject and contents might improve search results. This paper presents a novel method of document clustering, which uses semantic clique. First, we extracted the Features from the documents. Later, the associations between frequently co-occurring terms were defined, which were called as semantic cliques. Each connected component in the semantic clique represented a theme. The documents clustered based on the theme, for which we designed an aggregation algorithm. We evaluated the aggregation algorithm effectiveness using four kinds of datasets. The result showed that the semantic clique based document clustering algorithm performed significantly better than traditional clustering algorithms such as Principal Direction Divisive Partitioning (PDDP), k-means, Auto-Class, and Hierarchical Clustering (HAC). We found that the Semantic Clique Aggregation is a potential model to represent association rules in text and could be immensely useful for automatic document clustering.

Research on multi-document summarization based on latent semantic indexing

A multi-document summarization method based on Latent Semantic Indexing (LSI) is proposed. The method combines several reports on the same issue into a matrix of terms and sentences, and uses a Singular Value Decomposition (SVD) to reduce the dimension of the matrix and extract features, and then the sentence similarity is computed. The sentences are clustered according to similarity of sentences. The centroid sentences are selected from each class. Finally, the selected sentences are ordered to generate the summarization. The evaluation and results are presented, which prove that the proposed methods are efficient.

Designing a Document Retrieval Method for University Digital Libraries Based on Hadoop Technology

With the development of big data,all walks of life in society have begun to venture into big data to serve their own enterprises and departments.Big data has been embraced by university digital libraries.The most cumbersome work for the management of university libraries is document retrieval.This article uses Hadoop algorithm to extract semantic keywords and then calculates semantic similarity based on the literature retrieval keyword calculation process.The fast-matching method is used to determine the weight of each keyword,so as to ensure an efficient and accurate document retrieval in digital libraries,thus completing the design of the document retrieval method for university digital libraries based on Hadoop technology.

基于注意力机制语义增强的文档级关系抽取

文档级关系抽取旨在从文档中抽取出多个实体对之间的关系,具有较高的复杂性。针对文档级关系抽取中的多实体、关系相关性、关系分布不平衡等问题,提出了一种基于注意力机制(Attention)语义增强的文档级关系抽取方法,能够实现实体对之间关系的推理。具体来说,首先在数据编码模块改进编码策略,引入更多实体信息,通过编码网络捕获文档的语义特征,获得实体对矩阵;然后,设计了一个基于Attention门控机制的U-Net网络,对实体对矩阵进行局部信息捕获和全局信息汇总,实现语义增强;最后,使用自适应焦点损失函数缓解关系分布不平衡的问题。在4个公开的文档级关系抽取数据集(DocRED、CDR、GDA和DWIE)上评估了Att-DocuNet模型并取得了良好的实验结果。

组合光透视下小面积室内封闭空间分割方法

为提高室内空间的优化规划设计能力,提出基于点云数据语义分割的组合光透视下小面积室内封闭空间分割方法。构建小面积室内封闭空间的环境三维信息感知模型,采用室内空间点云提取室内封闭空间图像的坐标信息,结合语义组合特征分割方法将融合空间信息映射到高分辨空间异质单元结构中,引入几何约束的子空间投影特征信息,结合组合光透视的高分辨率分割图模型参数融合方法,提取小面积室内封闭的端元,采用点云数据语义分割方法实现空间分割。结果表明,采用该方法有效实现对复杂室内场景的三维重建,空间分割的均方根误差较低为0.808%,峰值信噪比较高42.156 dB,空间分割速度较快,平均为12.83 ms。

基于改进U-Net的车间场景分割

为了更好地对车间环境进行把控,同时针对车间目标像素级分割任务样本少、类别多、尺度变化大导致的语义分割精度低的问题,设计了一种改进的U-Net车间场景分割模型。改进的模型采用Rep-VGG轻量级主干网络,并在U-Net上采样阶段引入金字塔拆分注意力机制,以增加模型的特征表达能力及推理速度。模型训练则采用Dice-Cross Entropy组合损失函数以解决样本不均衡导致的难以训练的问题。实验数据表明该模型在自建的小样本车间数据集上可达到快速、轻量化及高精度的分割效果。

名词融合与复杂谓词的形成

文章以及物化的“VN”型复杂谓词(“踩雷乐视、取道北京”)为出发点,引入跨语言视角,探索汉语名词融合的句法、语义机制以及类型特征。在句法上,“V”和“N”通过词根移位融合,然后和事件有关的其他论旨角色会提升到“N”原有的论元槽位,表现为“VN”结构携带旁格论元的情况。在语义上,“N”并非“V”独立的语义论元,而是对“V”的语义进行修饰,对“V”所述事件进行限定和细分,但不改变“V”的价位。及物化的“VN”型复杂谓词的产生,不仅证明汉语中存在丰富的名词融合现象,更表明名词融合内部存在不同的演化轨迹,呈现谱系特性。文章提出的将形式语义学与分布式形态学相结合的形态—语义界面思路,为处理汉语中次短语层面的语言现象提供了新的框架性构想。

试析医养结合规范性文件的法理依据、效力及合法性要求

2013年国家提出医养结合新型养老模式,是应对人口老龄化的创新举措。中央和省级政府及其部门陆续出台一系列规范性文件,促进各地探索医养结合体制机制试点工作,并取得一定成效。然而,由于这些规范性文件效力位阶低,其创设的相关义务超越所涉上位法规定,致使医养结合面临制度性障碍。对此,应加大力度推进医养结合法治建设,为医养结合发展提供法治保障。这也是我国全面依法治国战略对医养结合依法推进的必然要求。

基于自然语言语义分析的Internet文件分类与过滤

网上文件过滤是网络信息安全研究的重要课题。传统的过滤方法简单地把关键字匹配作为分类的依据 ,常导致漏判误判等问题。文中提出一种基于自然语言理解的语义模板算法解决网上文件分类过滤的问题。实验结果表明该方法漏判误判率较低 。

Windows工作环境下VC^(++)与Matlab的联合应用研究

在分析 Visual C+ + 6.0和 Matlab5 .3各自提供的程序接口的基础上 ,提出了一种通过数据文件将 Visual C+ + 和 Matlab联合在一起进行软件开发的方法 ,并通过一个实例表明该方法适合于开发一个在 Windows工作环境下运行的、具有丰富的程序接口和良好的用户界面 ,以及复杂计算能力的应用软件 ,且该软件具有高效、易扩充等特点 .

OFD版式文档国家标准解读

分析了OFD在电子文件标准体系中的定位,重点阐述了OFD的成像模型技术、自定义标引技术、安全支撑、多文档等方面的重要特性,并给出了OFD在电子公文、档案、电子证照等领域的应用实例。

基于异构图分层学习的细粒度多文档摘要抽取

抽取的目标是在多个文档中提取共有关键信息,其对简洁性的要求高于单文档摘要抽取。现有的多文档摘要抽取方法通常在句子级别进行建模,容易引入较多的冗余信息。为了解决上述问题,提出一种基于异构图分层学习的多文档摘要抽取框架,通过层次化构建单词层级图和子句层级图来有效建模语义关系和结构关系。针对单词层级图和子句层级图这2个异构图的学习问题,设计具有不同层次更新机制的两层学习层来降低学习多种结构关系的难度。在单词层级图学习层,提出交替更新机制更新不同的粒度节点,以单词节点为载体通过图注意网络进行语义信息传递;在子句层级图学习层,提出两阶段分步学习更新机制聚合多种结构关系,第一阶段聚合同构关系,第二阶段基于注意力聚合异构关系。实验结果表明,与抽取式基准模型相比,该框架在Multinews数据集上取得了显著的性能提升,ROUGE-1、ROUGE-2和ROUGE-L分别提高0.88、0.23和2.27,消融实验结果也验证了两层学习层及其层次更新机制的有效性。

基于改进TF-IDF与BERT的领域情感词典构建方法

领域情感词典的构建是领域文本情感分析的基础。现有的领域情感词典构建方法存在所筛选候选情感词冗余度高、情感极性判断失准、领域依赖性强等问题。为了提高所筛选候选情感词的领域性和判断领域情感词极性的准确程度,提出了一种基于改进词频-逆文档频率(TF-IDF)与BERT的领域情感词典构建方法。该方法在筛选领域候选情感词阶段对TF-IDF算法进行改进,将隐含狄利克雷分布(LDA)算法与改进后的TF-IDF算法结合,进行领域性修正,提升了所筛选候选情感词的领域性;在候选情感词极性判断阶段,将情感倾向点互信息算法(SO-PMI)与BERT结合,利用领域情感词微调BERT分类模型,提高了判断领域候选情感词情感极性的准确程度。在不同领域的用户评论数据集上进行实验,结果表明,该方法可以提高所构建领域情感词典的质量,使用该方法构建的领域情感词典用于汽车领域和手机领域文本情感分析的F1值分别达到78.02%和88.35%。

基于语义排序功能实现对比文件快速筛选

为提高专利的审查质量和效率,本文提出了在智能检索系统下灵活运用语义排序功能以实现对比文件的快速筛选。通过结合一个实际案例的检索分析,阐述了如何在检索结果中运用语义排序功能。采用申请号或原权利要求作为一次语义排序基准,并以发明构思作为二次语义排序基准,可以将目标文献的排序提前,有效提高筛选效率。

多尺度特征融合的版面分析方法

针对当前文档版面元素分析中存在的列表和文本错分,表格内小尺度文本难以识别以及空间特征保留效果差等问题,本文基于自底向上的思想,提出一种基于SegNet网络的多特征融合版面分析方法。本文方法在SegNet中引入MSCAN-SE模块,针对表格中的小尺度元素识别率低的问题,利用注意力机制MSCAN-SE中的条状特征来提升模型多尺度特征的提取能力,使得网络能够保留更多尺度的特征信息;针对列表元素和文本元素特征过于相似的问题,通过注意力机制MSCAN-SE中的空洞卷积以及通道注意力分支来扩大网络在特征提取过程的感受野。本文方法与经典的语义分割网络通过实验进行性能比较,结果表明:本文方法在版面分析的测试集上的像素准确率为97.9%,平均交并比为91.7%,平均交并比较U-Net语义分割模型、FCN语义分割模型、DeepLabV3+语义分割模型和SegNet语义分割模型分别提高了7.6%、2.4%、2.6%和1.5%。

基于堆栈集成学习的文档隐含语义相似度判断算法

受到文档数量规模的影响,文档隐含语义相似度判断结果存在不精准的问题,提出基于堆栈集成学习的文档隐含语义相似度判断算法。构建堆栈集成学习模型,输入文档隐含语义,获得与原文本一一对应的原始文本词袋向量。赋值名词、副词、形容词,借助辅助词库矢量,识别类似隐含语义。构建文档隐含语义向量模型,提取基于文档隐含语义索引关键词。使用堆栈集成学习方法训练关键词,计算相似度获取判别结果。实验结果表明,该算法与实际关键词提取个数最大误差为1个,最高召回率为86%,相似度判断结果较为精准。