Semantic information integration and question answering based on pervasive agent ontology

The traditional search engines return a large number of relative web pages rather than accurate answers. However, in a question answering system, users could use sentences in daily life to raise questions. The question answering system will analyze and comprehend these questions and return answers to users directly. Aiming at the problems in current network environment, such as low precision of question answering, imperfect expression of domain knowledge, low reuse rate and lack of reasonable theory reference models, we put forward the information integration method of semantic web based on pervasive agent ontology (SWPAO) method, which will integrate, analyze and process enormous web information and extract answers on the basis of semantics. With SWPAO method as the clue, we mainly study the method of concept extraction based on uniform semantic term mining, pervasive agent ontology construction method on account of multi-points and the answer extraction in view of semantic inference. Meanwhile, we present the structural model of the question answering system applying ontology, which adopts OWL language to describe domain knowledge base from where it infers and extracts answers by Jena inference engine, thus the precision of question answering in QA system could be improved. In the system testing, the precision has reached 86%, and recalling rate is 93%. The experiment indicates that this method is feasible and it has the significance of reference and value of further study for the question answering systems.     

Question answering based on pervasive agent ontology and Semantic Web

Semantic Web technologies bring new benefits to knowledge-based question answering system. Especially, ontology is becoming the pivotal methodology to represent domain-specific conceptual knowledge in order to promote the semantic capability of a QA system. In this paper we present a QA system in which the domain knowledge is represented by means of ontology. In addition, personalized services are enabled through modeling users’ profiles in the form of pervasive agent ontology, and a Chinese Natural Language human–machine interface is implemented mainly through a NL parser in this system. An initial evaluation result shows the feasibility to build such a semantic QA system based on pervasive agent ontology, the effectivity of personalized semantic QA, the extensibility of pervasive agent ontology and knowledge base, and the possibility of self-produced knowledge-based on semantic relations in the pervasive agent ontology.     

基于本体的自动答疑系统的研究与实现

针对现有自动答疑系统中知识表示的不足,提出了用本体构建课程领域知识库的方案。根据答疑问句的特点进行了问题分类,将用户问句意图转换为对本体中基本元素的查询,并通过Jena语句加以实现,最终抽取出答案,实验证明方案是可行的。     

网络答疑本体的生成与匹配方法研究*

针对网络答疑具体任务,提出答疑本体概念,构建答疑本体,并提出一种基于答疑本体的答案匹配方法和相似度计算公式。公式不仅考虑词之间的相似度,还考虑问题答案对(QAp)在本体中的位置。该位置信息隐含了句子结构上的语义和词的语义扩展。实验结果表明提出的答疑本体能够有效表示QAp,易于答疑系统的检索匹配。     

一种基于形式概念分析的问答系统答案抽取的研究

为了在问答系统中尝试回答更加复杂的问题,有必要存在一种原理性的方法来动态产生不同的问答策略。提出在问答系统中使用形式概念分析（FCA）来抽取答案。在抽取过程中,首先在常问问题集（FAQs）中寻找已经存在的问题,如果在FAQs中的答案不满足用户的需求,再通过搜索引擎获取相关的文档。接着利用这些文档中前N个文档构建概念格,进而采用概念匹配在格中抽取答案。对于不同的问题,使用不同的策略进行匹配和抽取答案。     

基于BERT的三阶段式问答模型

预训练语言模型的发展极大地推动了机器阅读理解任务的进步.为了充分利用预训练语言模型中的浅层特征,并进一步提升问答模型预测答案的准确性,提出了一种基于BERT的三阶段式问答模型.首先,基于BERT设计了预回答、再回答及答案调整三个阶段;然后,在预回答阶段将BERT嵌入层的输入视作浅层特征来进行答案预生成;接着,在再回答阶...     

一种基于本体的辅助诊断治疗方法

以医学领域本体为基础,为了在医疗领域获取知识优势,结合机器学习技术,采用基于信息熵的属性加权方法结合基于实例的方法对病例进行分类,提出了一种基于本体辅助诊断治疗的方法。对相关工作进行了分析比较,给出了一个辅助诊断治疗心肌梗塞的例子,以说明该方法的有效性。与一般的医疗诊断方法相比,具备更良好的共享能力和扩展能力,可提高辅助医疗诊断、治疗的准确性。最后给出了结论和下一步工作。     

基于Chunk-CRF的情感问答研究

相对于事实性问答系统而言,观点或情感问答系统的研究除了需要考虑观点持有者及情感倾向性等与情感相关问题以外,其难点还在于答案形式更复杂更分散.从百度知道人工搜集了大量的情感问题,并根据情感问题的特征,统计并归纳了五大情感问题类型.问题分类模式与传统事实性问答系统不同,不能仅仅根据疑问词对其进行分类,还需要考虑到观点以及受众的反应.问题分类使用基于Chunk的CRF模型与规则相结合的情感问题分类方法.在答案抽取时结合组块识别的结果和情感的倾向性,并根据情感问题类型的不同采取不同的方法以获取答案.实验结果表明了评价体系的有效性.     

Semantic passage segmentation based on sentence topics for question answering

We propose a semantic passage segmentation method for a Question Answering (QA) system. We define a semantic passage as sentences grouped by semantic coherence, determined by the topic assigned to individual sentences. Topic assignments are done by a sentence classifier based on a statistical classification technique, Maximum Entropy (ME), combined with multiple linguistic features. We ran experiments to evaluate the proposed method and its impact on application tasks, passage retrieval and template-filling for question answering. The experimental result shows that our semantic passage retrieval method using topic matching is more useful than fixed length passage retrieval. With the template-filling task used for information extraction in the QA system, the value of the sentence topic assignment method was reinforced.     

Discourse processing for context question answering based on linguistic knowledge

Motivated by the recent effort on scenario-based context question answering (QA), this paper investigates the role of discourse processing and its implication on query expansion for a sequence of questions. Our view is that a question sequence is not random, but rather follows a coherent manner to serve some information goals. Therefore, this sequence of questions can be considered as a mini discourse with some characteristics of discourse cohesion. Understanding such a discourse will help QA systems better interpret questions and retrieve answers. Thus, we examine three models driven by Centering Theory for discourse processing: a reference model that resolves pronoun references for each question, a forward model that makes use of the forward looking centers from previous questions, and a transition model that takes into account the transition state between adjacent questions. Our empirical results indicate that more sophisticated processing based on discourse transitions and centers can significantly improve the performance of document retrieval compared to models that only resolve references. This paper provides a systematic evaluation of these models and discusses their potentials and limitations in processing coherent context questions.     

A novel approach to ontology classification

Ontology classification–the computation of the subsumption hierarchies for classes and properties–is a core reasoning service provided by all OWL reasoners known to us. A popular algorithm for computing the class hierarchy is the so-called Enhanced Traversal (ET) algorithm. In this paper, we present a new classification algorithm that attempts to address certain shortcomings of ET and improve its performance. Apart from classification of classes, we also consider object and data property classification. Using several simple examples, we show that the algorithms commonly used to implement these tasks are incomplete even for relatively weak ontology languages. Furthermore, we show that property classification can be reduced to class classification, which allows us to classify properties using our optimised algorithm. We implemented all our algorithms in the OWL reasoner HermiT. The results of our performance evaluation show significant performance improvements on several well-known ontologies.     

A reasoning enhance network for muti-relation question answering

Applied Intelligence - Multi-relation Question Answering is an important task of knowledge base over question answering (KBQA), multi-relation means that the question contains multiple relations...     

基于多路语义图网络的图像自动问答

基于视觉特征与文本特征融合的图像问答已经成为自动问答的热点研究问题之一。现有的大部分模型都是通过注意力机制来挖掘图像和问题语句之间的关联关系,忽略了图像区域和问题词在同一模态之中以及不同视角的关联关系。针对该问题,提出一种基于多路语义图网络的图像自动问答模型（MSGN）,从多个角度挖掘图像和问题之间的语义关联。MSGN利用图神经网络模型挖掘图像区域和问题词细粒度的模态内模态间的关联关系,进而提高答案预测的准确性。模型在公开的图像问答数据集上的实验结果表明,从多个角度挖掘图像和问题之间的语义关联可提高图像问题答案预测的性能。     

基于HNC句类的社区问答系统问句检索模型构建

社区问答系统中充斥着大量的噪声,给用户检索信息造成麻烦,以往的问句检索模型大多集中在词语层面。针对以上问题构建句子层面的问句检索模型。新模型基于概念层次网络（hierarchincal network of concept,HNC）理论当中的句类知识,从句子的语用、语法和语义三个层面计算问句间相似度。通过问句分类算法确定查询问句和候选问句的问句类别,得到问句间的语用相似度,利用句类表达式的结构和语义块组成分别计算问句间的语法及语义相似度。在真实数据集上的实验表明,基于HNC句类的新模型提高了问句检索结果的准确性。     

A question answering system for project management applications

The usage of computer applications in the construction industry is increasing, as is the complexity of software applications and this makes it difficult for project personnel to maintain familiarity. Furthermore, the causes of practical problems, such as project delays and cost over-runs, are often not derivable from the output of most software. A question answering system provides a means for directly extracting knowledge from this output. This paper begins with an examination of issues involved in building such a system. An emerging industry standard, ifcXML, is adopted as the knowledge representation format, thereby reducing the effort that is necessary to build a knowledge base. We then explore the mechanisms that use information in the knowledge base for question understanding. A prototype system has been built and tested to illustrate usefulness for project management applications.     

基于章节本体和转移网络的自动答疑系统

针对现有本体和转移网络在网络教育自动答疑系统中应用的不足,提出了一种基于章节划分的本体与转移网络方法。定义了章节本体,确定了基于章节本体的问题模式,给出了基于章节划分转移网络解决规范化问题的答案生成。基于章节本体知识给出了语义相似度模型,用于解决非规范问题的答案生成;基于章节划分转移网络建立了章节索引转移子网的搜索方法。实验结果表明,基于章节划分的本体论和转移网络方法在不降低搜索效果的情况下,比一般的本体论和转移网络方法的查询时间减少了38%。     

基于自动问答系统的信息检索技术研究进展

自动问答是根据用户以自然语言提出的问题给出一个明确的答案。近年来,自动问答越来越受到信息检索和自然语言处理的研究者的关注。典型的自动问答系统通常包含问题分析、文段检索和答案选择等部件。介绍了自动问答的最新研究进展和相关国际会议情况,着重阐述问题分类、查询扩展、文段检索和答案选择这四个热点技术的主要功能和常用方法,最后提出存在的一些问题和展望。     

基于互联网和self-training的中文问答模式学习

在已有的问答模式学习中,模式定义和候选答案评分偏于简单,而且学习过程依赖于人工标定语料。通过挖掘Web文本中动、名词序列的骨架模式,用以扩充模式定义;将self-training学习机制引入问答模式学习：用一对训练语料进行初始学习,通过互联网搜索,自动选择可靠程度较高的问答对,重新训练;扩充了启发规则,改进候选答案的评分方法。实验结果表明：所提出的问答模式学习方法能有效地提高中文问答系统的性能。     

基于多特征语义匹配的知识库问答系统

知识库问答(KBQA)任务主要目的在于精确地将自然语言问题和知识库(KB)中的三元组进行匹配。传统的KBQA方法通常专注于实体识别和谓语匹配,实体识别的错误会导致错误传播从而无法得到正确的答案。针对上述问题提出一种端到端的解决方案直接匹配问题和三元组,该系统主要包含候选三元组生成和候选三元组排序两个部分来实现精确问答。首先通过BM25算法计算问题和知识库中三元组的相关性生成候选三元组;然后通过多特征语义匹配模型(MFSMM)进行三元组的排序,即用MFSMM分别通过双向长短时记忆网络(Bi-LSTM)和卷积神经网络(CNN)实现语义相似度和字符相似度的计算,并通过融合来对三元组进行排序。该系统在NLPCC-ICCPOL 2016KBQA数据集上的平均F1为80. 35%,接近了现有最好的表现。     

基于子问题渐进式推理的3D视觉问答

3D视觉问答可以帮助人们理解空间信息,在幼儿教育等方面具有广阔的应用前景。3D场景信息复杂,现有方法大多直接进行回答,面对复杂问题时容易忽视上下文细节,从而导致性能下降。针对该问题,提出了一种基于子问题渐进式推理的3D视觉问答方法,通过文本分析为复杂的原始问题构建多个简单的子问题。模型在回答子问题的过程中学习上下文信息,帮助理解复杂问题的含义,最终利用积累的联合信息得出原始问题的答案。子问题与原始问题呈现渐近式推理关系,使得模型具有明确的错误解释性和可追溯性。在现有3D数据集ScanQA上进行的实验表明,所提方法在EM@10和CIDEr两个指标上分别达到了51.49%和61.68%,均超过了现有的其他3D视觉问答方法,证实了该方法的有效性。