Semantic Analysis of Natural Language Queries for an Object Oriented Database

This paper presents the semantic analysis of queries written in natural language (French) and dedicated to the object oriented data bases. The studied queries include one or two nominal groups (NG) articulating around a verb. A NG consists of one or several keywords (application dependent noun or value). Simple semantic filters are defined for identifying these keywords which can be of semantic value: class, simple attribute, composed attribute, key value or not key value. Coherence rules and coherence constraints are introduced, to check the validity of the co-occurrence of two consecutive nouns in complex NG. If a query is constituted of a single NG, no further analysis is required. Otherwise, if a query covers two valid NG, it is a subject of studying the semantic coherence of the verb and both NG which are attached to it.

Query Expansion Based on Semantics and Statistics in Chinese Question Answering System

In Chinese question answering system, because there is more semantic relation in questions than that in query words, the precision can be improved by expanding query while using natural language questions to retrieve documents. This paper proposes a new approach to query expansion based on semantics and statistics Firstly automatic relevance feedback method is used to generate a candidate expansion word set. Then the expanded query words are selected from the set based on the semantic similarity and seman- tic relevancy between the candidate words and the original words. Experiments show the new approach is effective for Web retrieval and out-performs the conventional expansion approaches.

Evaluating Relational Ranking Queries Involving both Text Attributes and Numeric Attributes

In many database applications, ranking queries may reference both text and numeric attributes, where the ranking functions are based on both semantic distances/similarities for text attributes and numeric distances for numeric attributes. In this paper, we propose a new method for evaluating such type of ranking queries over a relational database. By statistics and training, this method builds a mechanism that combines the semantic and numeric distances, and the mechanism can be used to balance the effects of text attributes and numeric attributes on matching a given query and tuples in database search. The basic idea of the method is to create an index based on WordNet to expand the tuple words semantically for text attributes and on the information of numeric attributes. The candidate results for a query are retrieved by the index and a simple SQL selection statement, and then top-N answers are obtained. The results of extensive experiments indicate that the performance of this new strategy is efficient and effective.

Applying First Order Predicate to the Semantic Query of Chinese

This paper described an approach to make inferences on Chinese information using first order predicate logic, which could be used in the semantic query of Chinese. The predicates of the method were derived from the natural language using rule based LFT, the axiom set was generated by extracting lexicon knowledge from HowNet, and the first order predicate inferences were made through symbol connection of center words. After all these were done, the evaluation and possible improvements of the method were provided. The experiment result shows a higher precision rate than that traditional methods can reach.

An Efficient Algorithm for Query Transformation in Semantic Query Optimization

Semantic query optimization (SQO) is comparatively a recent approach for the transformation of given query into equivalent alternative query using matching rules in order to select an optimal query based on the costs of executing alternative queries. The key aspect of the algorithm proposed here is that previous proposed SQO techniques can be considered equally in the uniform cost model, with which optimization opportunities will not be missed. At the same time, the authors used the implication closure to guarantee that any matched rule will not be lost. The authors implemented their algorithm for the optimization of decomposed sub-query in local database in Multi-Database Integrator (MDBI), which is a multidatabase project. The experimental results verify that this algorithm is effective in the process of SQO.

不确定性Top-K查询处理

高效Top-K查询处理在涉及大量数据交互的应用中是一项重要技术,随着应用中不确定性数据的大量涌现,不确定性数据的管理逐渐引起人们的重视.不确定性数据上Top-K查询从语义和处理上都呈现出与传统Top-K查询不同的特点.在主流不确定性数据模型和可能世界语义模型下,学者们已经提出了多种不确定性Top-K查询的语义和处理方法.介绍了当前不确定性Top-K查询的研究工作,并对其进行分类,讨论包括语义、排序标准、算法以及应用等方面的技术.最后提出不确定性Top-K查询面临的挑战和下一步的发展方向.

基于Web的数字化资源共享信息查询系统设计

数字化资源增多使得共享信息查询难度增加,为此,设计了基于Web的数字化资源共享信息查询系统。通过统一用户接口,实现数字化资源共享信息的交互和集成,构建查询系统管理平台。连接外部共享信息提供程序和服务执行个体,利用Web服务器的索引服务,完成信息局部查询。管理远程信息查询节点服务的请求,改变创建端点的新路由,防止信息丢失,结合语义相关优化系统,根据关键字的反向资源频次,获得辨识能力,实现数字化资源共享信息全局查询。实验结果表明,基于Web的信息查询系统的查询处理能力较强,能够提高查询准确率和查询效率。

基于依存句法分析的数学表达式查询扩展方法

传统的数学表达式检索方法主要面向表达式的二维结构,难以检索出具有相同语义但结构不同的数学表达式。针对这一问题,设计一种基于依存句法分析的数学表达式查询扩展方法。以FDS(Formula Description Structure)解析表达式结构信息的检索方法为基础,通过依存句法抽取表达式周围文本中的语义词,并建立运算符、表达式、语义词间索引;通过语义词二次查询,实现检索语义相同但结构不同数学表达式的目的。实验结果表明,依存分析能有效地抽取数学表达式语义词,加入语义的表达式检索方法,查全率和查准率都有了一定的提高。

支持OR语义的高效受限Top-k空间关键字查询技术

近些年,随着定位系统和移动设备的普及,空间文本对象的数量日益庞大,基于位置的地理信息服务在人们的生活中发挥着越来越重要的作用.对于空间关键字查询搜索的研究亦如火如荼.然而,现有许多研究工作只适用于AND语义,支持OR语义的搜索研究相对较少.当用户放松对关键字匹配的要求时,支持OR语义的搜索技术显得尤为重要.针对这一问题,在聚集线性四分树的基础上,利用线性四分树上物理存储的Morton码与逻辑空间位置的对应性,提出了基于虚拟网格的VGrid算法.该算法可同时支持OR语义和AND语义.最后,通过在真实数据集上进行大量实验,验证了所提算法的有效性和高效性.

抵御背景信息推理攻击的假位置生成算法

针对已有的假位置生成算法,设计了一种多次查询请求攻击算法(Multiple Query Request Attack algorithm,MQRA)来测试其安全性。为有效保护用户的位置隐私,提出了一种抵御背景信息推理攻击的假位置生成算法(Dummy Location Ge-neration Algorithm against Side Information Inference Attack,DLG_SIA),该算法综合考虑了查询概率、时间分布、位置语义和物理分散度等背景信息来生成有效的假位置集以抵御概率分布攻击、位置语义攻击和位置同质攻击,避免攻击者结合背景信息过滤掉假位置。用户首次请求时,DLG_SIA算法先利用位置熵和时间熵选取当前请求时间下查询概率相似的位置点来生成假位置集,并通过调整的余弦相似度生成满足语义差异性的位置点;然后通过距离熵保证选取的位置点间具有更大的匿名范围,并将当前请求位置的最佳假位置集进行缓存。安全性分析和仿真实验结果表明:MQRA算法能以很高的概率识别出假位置集中用户的真实位置;与已有的假位置生成算法相比,DLG_SIA算法能有效抵御背景信息推理攻击,保护用户的位置隐私。

基于混合神经网络的自然语言处理技术研究

针对当前电网运营数据指标查询便捷性较差的问题,提出了基于BERT的电力指标查询统计专家库模型。该模型采用BERT语言建模方法对电力指标查询统计文本进行了处理,并将其转化为由词向量、句向量及位置向量叠加形成的输入文本向量。同时引入BERT预训练过程来实现上下文依赖关系,以进行初步挖掘。利用基于Transformer模型的深度语义识别模型,通过BERT精调过程深入挖掘输入文本与语义识别结果之间的内在联系。构建了涵盖准确率、召回率的评价指标来评估算法的性能。算例分析结果表明,相比于RNN-LSTM和BiLSTM-Attention模型,所提模型的评价指标更高且计算时间较少,故在准确性与计算速度方面均存在优势;同时将该文所提BERT模型应用于实际电力指标查询文本中,结果发现工作人员对线损率指标尤为关注,应将其作为电力企业运营评估的重要工程指标。

Supporting Various Top-k Queries over Uncertain Datasets

There have been many researches and semantics in answering top-k queries on uncertain data in various applications. However, most of these semantics must consume much of their time in computing position probability. Our approach to support various top-k queries is based on position probability distribution （PPD） sharing. In this paper, a PPD-tree structure and several basic operations on it are proposed to support various top-k queries. In addition, we proposed an approximation method to improve the efficiency of PPD generation. We also verify the effectiveness and efficiency of our approach by both theoretical analysis and experiments.

Answering Non-Answer Questions on Reverse Top-k Geo-Social Keyword Queries

Due to the wide-spread use of geo-positioning technologies and geo-social networks,the reverse top-k geo-social keyword query has attracted considerable attention from both industry and research communities.A reverse top-k geo-social keyword(RkGSK)query finds the users who are spatially near,textually similar,and socially relevant to a specified point of interest.RkGSK queries are useful in many real-life applications.For example,they can help the query issuer identify potential customers in marketing decisions.However,the query constraints could be too strict sometimes,making it hard to find any result for the RkGSK query.The query issuers may wonder how to modify their original queries to get a certain number of query results.In this paper,we study non-answer questions on reverse top-k geo-social keyword queries(NARGSK).Given an RkGSK query and the required number M of query results,NARGSK aim to find the refined RkGSK query having M users in its result set.To efficiently answer NARGSK,we propose two algorithms(ERQ and NRG)based on query relaxation.As this is the first work to address NARGSK to the best of our knowledge,ERQ is the baseline extended from the state-of-the-art method,while NRG further improves the efficiency of ERQ.Extensive experiments using real-life datasets demonstrate the efficiency of our proposed algorithms,and the performance of NRG is improved by a factor of 1–2 on average compared with ERQ.

P2P数据管理

P2P(peer-to-peer)技术是未来重构分布式体系结构的关键技术,拥有广阔的应用前景.P2P系统的大多数问题都可归结为数据放置和检索问题,因此,P2P数据管理成为数据库领域活跃的研究课题.当前,P2P数据管理主要有信息检索、数据库查询和连续查询3个子领域,取得了许多研究成果.在介绍P2P技术的优点后,指出了P2P数据管理研究的目标.然后针对上述3个方面,论述P2P数据管理研究的现状,着重讨论了P2P数据库查询的索引构造策略、语义异构的解决方法、查询语义、查询处理策略、查询类型和查询优化技术.通过比较,指出了现状与目标的差距,提出了需要进一步研究的问题.

基于地理空间本体的语义检索相关度研究

以地理信息领域为应用背景,面向地理空间语义检索,基于地球信息科学中的空间拓扑理论,以空间本体为语义检索的概念空间,提出了一种语义相关度的算法。其特点是考虑了传统字面匹配相关度与语义关系相关度两部分的融合,同时引入了本体关系权值的机制控制在不同语义检索应用中本体的关联程度,并体现了其与语义距离的反比关系。通过所作的相关实验,验证了该语义相关度算法在地理空间语义检索应用中可以达到良好的效果,并且也为其他领域应用提供了较好的参考和借鉴价值。

基于语义相似度的数据库自适应查询松弛方法

为了解决数据库空查询结果问题,提出了一种基于语义相似度的数据库自适应查询松弛方法.首先,基于初始查询条件和数据分布推测用户对查询指定属性的重视程度,据此提出了一种属性权重评估方法;然后,通过考察属性值的特征信息,分别提出了分类型属性值之间和数值型属性值之间的语义相似度评估方法;在此基础上,根据松弛阈值、属性权重和属性值之间的语义相似度,提出了一种自适应查询松弛重写算法.对于满足松弛查询的近似查询结果,按照它们对初始查询的满足度进行排序.实验结果表明,提出的属性权重和属性值之间的语义相似度评估方法性能稳定,评估结果合理,提出的查询松弛方法具有较高的查全率,能够较好地满足当前用户的需求和偏好.

查询扩展技术进展与展望

查询扩展技术是改善信息检索中查全率和查准率的关键技术之一,并成为近年来研究的热点。以查询扩展技术的发展为主线,将目前的研究方法归纳为三类:关键词查询扩展技术、查询词权重调整技术和语义概念查询扩展技术。对这三类扩展技术进行了整体性阐述和比较性研究,对一些关键性问题进行了探讨,并展望未来的发展趋势。

基于语义关系查询扩展的文档重构方法

已知文档与用户查询之间相同概念不同表达形式造成的词不匹配问题是影响信息检索效果的重要原因之一 .该文提出了根据词之间的语义关系进行扩展和替换的文档重构方法 .它与传统的查询扩展不同 ,实现了同一概念信息的聚集 ,是更接近于人类进行信息查找的思维方法 .进一步地 ,研究给出一种有效的实时文档重构检索策略 ,解决了文档重构方法在实际应用中的可行性 .在标准测试数据集上的实验表明 ,基于查询扩展的文档重构方法不仅比不扩展的最佳性能始终有 14 %～ 2 3.4 %的提高 ,而且比相对应的传统查询扩展方法也有约 16 %的提高 .

领域本体中的概念相似度计算

借鉴计算语言学中的语义距离思想,提出了RDFSchema构词所描述的本体概念相似度计算方法,并利用该方法对农业知识本体(AO)所描述的部分概念进行了相似度计算和分析.结果表明,该方法可以定量地分析概念、特性之间的相似度,并可以指导基于领域知识本体的语义查询中的概念集扩充和查询结果排序.

高可扩展的RDF数据存储系统

由于资源描述框架(resource description framework,RDF)具有表达灵活、简洁等优点,已被接受为表达元数据及万维网上数据互联的规范.近年来,其数据量在以飞快的速度增长.相应地,要求存储RDF数据的系统应具有高扩展性.介绍了一个高可扩展的RDF数据存储系统TripleBit.为尽可能降低存储空间消耗,采用了增量压缩和变长整数编码方法.并采用了数据分块的存储方法,既使得存储管理方便又使得存储结构紧凑,加速了数据读取.系统提供了基于启发式规则的动态查询计划生成方法,所产生的查询计划在执行过程中根据中间结果会相应作调整,以保持最优的执行顺序.对于多变量的查询,使用二步执行策略以减少查询过程中产生的中间结果.与目前流行RDF数据存储系统相比较,在存储空间上RDF-3X比TripleBit至少多40%;在查询性能上,比RDF-3X和MonetDB获得数倍的提升.