Similarity Intelligence:Similarity Based Reasoning,Computing,and Analytics

Similarity has been playing an important role in computer science,artificial intelligence(AI)and data science.However,similarity intelligence has been ignored in these disciplines.Similarity intelligence is a process of discovering intelligence through similarity.This article will explore similarity intelligence,similarity-based reasoning,similarity computing and analytics.More specifically,this article looks at the similarity as an intelligence and its impact on a few areas in the real world.It explores similarity intelligence accompanying experience-based intelligence,knowledge-based intelligence,and data-based intelligence to play an important role in computer science,AI,and data science.This article explores similarity-based reasoning(SBR)and proposes three similarity-based inference rules.It then examines similarity computing and analytics,and a multiagent SBR system.The main contributions of this article are:1)Similarity intelligence is discovered from experience-based intelligence consisting of data-based intelligence and knowledge-based intelligence.2)Similarity-based reasoning,computing and analytics can be used to create similarity intelligence.The proposed approach will facilitate research and development of similarity intelligence,similarity computing and analytics,machine learning and case-based reasoning.

A new similarity computing method based on concept similarity in Chinese text processing

The paper proposes a new text similarity computing method based on concept similarity in Chinese text processing. The new method converts text to words vector space model at first, and then splits words into a set of concepts. Through computing the inner products between concepts, it obtains the similarity between words. The new method computes the similarity of text based on the similarity of words at last. The contributions of the paper include： 1） propose a new computing formula between words; 2） propose a new text similarity computing method based on words similarity; 3） successfully use the method in the application of similarity computing of WEB news; and 4） prove the validity of the method through extensive experiments.

Design of ontology mapping framework and improvement of similarity computation

Ontology heterogeneity is the primary obstacle for interoperation of ontologies. Ontology mapping is the best way to solve this problem. The key of ontology mapping is the similarity computation. At present, the method of similarity computation is imperfect. And the computation quantity is high. To solve these problems, an ontology-mapping framework with a kind of hybrid architecture is put forward, with an improvement in the method of similarity computation. Different areas have different local ontologies. Two ontologies are taken as examples, to explain the specific mapping framework and improved method of similarity computation. These two ontologies are about classes and teachers in a university. The experimental results show that using this framework and improved method can increase the accuracy of computation to a certain extent. Otherwise, the quantity of computation can be decreased.

Design and Implementation of FAQ Automatic Return System Based on Similarity Computation

FAQ （frequently asked question） is widely used on the Internet, but most FAQ＇s asking and answering are not automatic. This paper introduces the design and imple mentation of a FAQ automatic return system based on semantic similarity computation, including computation model choo sing, FAQ characters analyzing, FAQ data formal expressing, feature vector indexing, and weight computing and so on. According to FAQ features of sentence length short, two mapping, strong domain characteristics etc. Vector Space Model with special semantic process was selected in system, and corresponding algorithm of similarity computation was proposed too. Experiment shows that the system has a good performance for high frequent and common questions.

Hadoop-Based Similarity Computation System for Composed Documents

There exist a large number of composed documents in universities in the teaching process. Most of them are required to check the similarity for validation. A kind of similarity computation system is constructed for composed documents with images and text information. Firstly, each document is split and outputs two parts as images and text information. Then, these documents are compared by computing the similarities of images and text contents independently. Through Hadoop system, the text contents are easily and quickly separated. Experimental results show that the proposed system is efficient and practical.

Information mining and similarity computation for semi-/un-structured sentences from the social data

In recent years,with the development of the social Internet of Things(IoT),all kinds of data accumulated on the network.These data,which contain a lot of social information and opinions.However,these data are rarely fully analyzed,which is a major obstacle to the intelligent development of the social IoT.In this paper,we propose a sentence similarity analysis model to analyze the similarity in people’s opinions on hot topics in social media and news pages.Most of these data are unstructured or semi-structured sentences,so the accuracy of sentence similarity analysis largely determines the model’s performance.For the purpose of improving accuracy,we propose a novel method of sentence similarity computation to extract the syntactic and semantic information of the semi-structured and unstructured sentences.We mainly consider the subjects,predicates and objects of sentence pairs and use Stanford Parser to classify the dependency relation triples to calculate the syntactic and semantic similarity between two sentences.Finally,we verify the performance of the model with the Microsoft Research Paraphrase Corpus(MRPC),which consists of 4076 pairs of training sentences and 1725 pairs of test sentences,and most of the data came from the news of social data.Extensive simulations demonstrate that our method outperforms other state-of-the-art methods regarding the correlation coefficient and the mean deviation.

Four-Dimensional Computerized Tomography (4D-CT) Reconstruction Based on the Similarity Measure of Spatial Adjacent Images

Objective:To investigate the feasibility of a 4D-CT reconstruction method based on the similarity principle of spatial adjacent images and mutual information measure. Methods:A motor driven sinusoidal motion platform made in house was used to create one-dimensional periodical motion that was along the longitudinal axis of the CT couch. The amplitude of sinusoidal motion was set to an amplitude of ±1 cm. The period of the motion was adjustable and set to 3.5 s. Phantom objects of two eggs were placed in a Styrofoam block, which in turn were placed on the motion platform. These objects were used to simulate volumes of interest undergoing ideal periodic motion. CT data of static phantom were acquired using a multi-slice general electric (GE) LightSpeed 16-slice CT scanner in an axial mode. And the CT data of periodical motion phantom were acquired in an axial and cine-mode scan. A software program was developed by using VC++ and VTK software tools to resort the CT data and reconstruct the 4D-CT. Then all of the CT data with same phase were sorted by the program into the same series based on the similarity principle of spatial adjacent images and mutual information measure among them, and 3D reconstruction of different phase CT data were completed by using the software. Results:All of the CT data were sorted accurately into different series based on the similarity principle of spatial adjacent images and mutual information measures among them. Compared with the unsorted CT data, the motion artifacts in the 3D reconstruction of sorted CT data were reduced significantly, and all of the sorted CT series result in a 4D-CT that reflected the characteristic of the periodical motion phantom. Conclusion:Time-resolved 4D-CT reconstruction can be implemented with any general multi-slice CT scanners based on the similarity principle of spatial adjacent images and mutual information measure.The process of the 4D-CT data acquisition and reconstruction were not restricted to the hardware or software of the CT scanner and has the feasibility ,which extensive applicability.

Multi-core based parallel computing technique for content-based image retrieval

In this paper, we propose a parallel computing technique for content-based image retrieval （CBIR） system. This technique is mainly used for single node with multi-core processor, which is different from those based on cluster or network computing architecture. Due to its specific applications （such as medical image processing） and the harsh terms of hardware resource requirement, the CBIR system has been prevented from being widely used. With the increasing volume of the image database, the widespread use of multi-core processors, and the requirement of the retrieval accuracy and speed, we need to achieve a retrieval strategy which is based on multi-core processor to make the retrieval faster and more convenient than before. Experimental results demonstrate that this parallel architecture can significantly improve the performance of retrieval system. In addition, we also propose an efficient parallel technique with the combinations of the cluster and the multi-core techniques, which is supposed to gear to the new trend of the cloud computing.

A New Method for Calculating Similarity between Sentences and Application on Automatic Abstracting

Sentence similarity computing plays an important role in machine question-answering systems, machine-translation systems, information retrieval and automatic abstracting systems. This article firstly sums up several methods for calculating similarity between sentences, and brings out a new method which takes all factors into consideration including critical words, semantic information, sentential form and sen-tence length. And on this basis, a automatic abstracting system based on LexRank algorithm is implemented. We made several improvements in both sentence weight computing and redundancy resolution. The system described in this article could deal with single or multi-document summarization both in English and Chinese. With evaluations on two corpuses, our system could produce better summaries to a certain degree. We also show that our system is quite insensitive to the noise in the data that may result from an imperfect topical clustering of documents. And in the end, existing problem and the developing trend of automatic summariza-tion technology are discussed.

Optimization method of Hadamard coding plate inγ‑ray computational ghost imaging

Owing to the constraints on the fabrication ofγ-ray coding plates with many pixels,few studies have been carried out onγ-ray computational ghost imaging.Thus,the development of coding plates with fewer pixels is essential to achieveγ-ray computational ghost imaging.Based on the regional similarity between Hadamard subcoding plates,this study presents an optimization method to reduce the number of pixels of Hadamard coding plates.First,a moving distance matrix was obtained to describe the regional similarity quantitatively.Second,based on the matrix,we used two ant colony optimization arrangement algorithms to maximize the reuse of pixels in the regional similarity area and obtain new compressed coding plates.With full sampling,these two algorithms improved the pixel utilization of the coding plate,and the compression ratio values were 54.2%and 58.9%,respectively.In addition,three undersampled sequences(the Harr,Russian dolls,and cake-cutting sequences)with different sampling rates were tested and discussed.With different sampling rates,our method reduced the number of pixels of all three sequences,especially for the Russian dolls and cake-cutting sequences.Therefore,our method can reduce the number of pixels,manufacturing cost,and difficulty of the coding plate,which is beneficial for the implementation and application ofγ-ray computational ghost imaging.

Gershgorin and Rayleigh Bounds on the Eigenvalues of the Finite-Element Global Matrices via Optimal Similarity Transformations

The large finite element global stiffness matrix is an algebraic, discreet, even-order, differential operator of zero row sums. Direct application of the, practically convenient, readily applied, Gershgorin’s eigenvalue bounding theorem to this matrix inherently fails to foresee its positive definiteness, predictably, and routinely failing to produce a nontrivial lower bound on the least eigenvalue of this, theoretically assured to be positive definite, matrix. Considered here are practical methods for producing an optimal similarity transformation for the finite-elements global stiffness matrix, following which non trivial, realistic, lower bounds on the least eigenvalue can be located, then further improved. The technique is restricted here to the common case of a global stiffness matrix having only non-positive off-diagonal entries. For such a matrix application of the Gershgorin bounding method may be carried out by a mere matrix vector multiplication.

A Fault Tolerance Algorithm for Resource Discovery in Semantic Grid Computing Using Task Agents

One of the interesting topics in grid computing systems is resources discovery. After the failure of a resource in a chain of resources made for a specific task in grid environment, discovering and finding a new resource that reconstructs the chain is an important topic. In this study, with defining new agent that is called task agent, and by proposing an algorithm, we will increase the fault tolerance against probable failure of a resource in the resource chain.

Illegal Access Detection in the Cloud Computing Environment

In this paper detection method for the illegal access to the cloud infrastructure is proposed. Detection process is based on the collaborative filtering algorithm constructed on the cloud model. Here, first of all, the normal behavior of the user is formed in the shape of a cloud model, then these models are compared with each other by using the cosine similarity method and by applying the collaborative filtering method the deviations from the normal behavior are evaluated. If the deviation value is above than the threshold, the user who gained access to the system is evaluated as illegal, otherwise he is evaluated as a real user.

Group Similarity and Social Influence Analysis in Online Communities

A fundamental open question in the analysis of social networks was to understand the evolution between similarity and group social ties.In general,two groups are similar for two distinct reasons:first,they grow to change their behaviors to the same group due to social influence;second,they tend to merge a group due to similar behaviors,where a process often is termed selection by sociologists.It was important to understand why two groups could merge and what led to high similarities for members in a group,influence or selection.In this paper,the techniques for identifying and modeling interactions between social influence and selection for different groups were developed.Different similarities were computed in three phases where groups came into being,before or after according to the number of common edits in Wikipedia.Experimental results showed selection played a more important role in two group merging.

基于特征相似性和特征规范化的注意力模块

近年来,注意力机制在图像分类、目标检测和语义分割等领域取得了巨大成功,但现有的注意力机制大多只能在通道或空间维度上实现特征融合,这极大限制了其在通道和空间维度上变化的灵活性,导致无法充分利用特征信息。为此,文中提出一种基于特征相似性和特征规范化的、可同时利用特征图各维度信息的卷积神经网络注意力模块FSNAM。该模块由特征相似性模块(FSM)和特征规范化模块(FNM)两部分组成,FSM利用输入特征图的通道特征信息和局部空间特征信息生成一个二维的特征相似性权重图;FNM利用输入特征图的全局空间特征信息生成一个三维的特征规范化权重图;两个模块生成的权重图融合在一起,生成一个三维的注意力权重图,以此实现通道特征信息和空间特征信息的融合。为证明FSNAM的可行性和有效性,进行了消融实验,结果表明:在图像分类任务方面,FSNAM模块对分类网络在CIFAR数据集上的性能提升明显优于其他主流注意力模块;在目标检测任务方面,使用FSNAM模块的目标检测网络对VOC数据集中的小目标和中等大小目标的检测准确率分别提高了3.9和1.2个百分点;在语义分割任务方面,使用FSNAM模块可以提高HRNet模型的性能,在SBD数据集上模型的平均像素准确率提高了0.58个百分点。

我国央地政府健康信息政策比较研究

[目的/意义]党的二十大提出把保障人民健康放在优先发展的战略地位,而政策作为实现这一目标的关键驱动力起到了重要的引导和支持作用。建立我国健康信息政策宏观与中观层面的联系并进行比较,能够推进我国健康事业的全面发展。[方法/过程]结合各地健康事业发展状况,利用文本相似度计算对国务院发布的《“十四五”国民健康规划》和我国25个地区发布的健康信息政策文本进行比较研究。[结果/结论]中医药健康服务、疫情救治、业态融合、老年人健康等内容在各地关注度最高,呈现出与中央的政策耦合。宁夏回族自治区及广西壮族自治区政策完整性强;安徽省、海南省逐渐形成健康产业区域特色;广东省、湖南省单项数值突出,呈现出政策制定的差异性。在此基础上,提出强化央地间的政策协同性、加强区域间的经验协作、挖掘区域优势构建特色健康产业格局的建议。

基于九屏幕法的计算机绘图相似辨识方法研究

为了制止学生使用AutoCAD绘图在各类教学中的诸多图形相似行为,在TRIZ的九屏幕法启发下,提出基于AutoCAD绘图过程的相似辨识方法。利用ObjectARX开发了学生端记录辨识图形相似的特征信息插件,基于Python、Django、MySQL等开发了B/S模式的教师端图形相似辨识系统。先通过九屏幕法对问题进行分析,得到解决问题的资源;其次,分析学生绘图过程的产生相似图形的绘图方式;再次,分析这些绘图方式所引起的特征信息变化;然后,利用开发的插件将特征信息存入数据库;最后,使用这些特征信息来进行相似图档的辨识。通过教学实践表明这套方法可以有效制止学生图形相似,显著提高计算机绘图的教学效果。

基于二级时空分桶的伴随轨迹查询

随着移动传感器设备的普及,人们能够采集到的位置数据越来越多,轨迹数据的规模也越来越庞大.从大规模时空数据中查找与指定轨迹最相似的前k条轨迹一直是时空大数据挖掘的重要挑战之一.现有的相似轨迹查询方法大都包括三个阶段:(1)对海量的离线轨迹数据建立索引;(2)基于索引结构从已知轨迹集中查询与指定轨迹相似的候选轨迹;(3)计算指定轨迹与候选轨迹之间的精确相似度并返回相似度最大的前k条轨迹.但大多数现有方法对轨迹进行聚类索引时不能有效利用时间和空间信息,导致时间相似度不高的轨迹也会被划分到相同的索引项上,最终影响查询的准确性和效率.此外,现有的时空轨迹相似度计算方法存在大量的无效运算,使得相似轨迹的查询效率整体较低.针对当前伴随轨迹查询方法对时间与空间信息利用不充分的问题,本文提出一种新的二级时空分桶索引结构,首先将每条轨迹数据按照时间滑动窗口划分为若干带有时间槽信息的子轨迹,在时间上对轨迹进行一级索引聚类;在此基础上对在相同时间槽内的子轨迹进行二级空间索引聚类,利用哈希算法将具有连续相同位置点的子轨迹映射到同一时空分桶中.与已有索引方法相比,该方法对不同轨迹在索引时具有更好的区分度,查询时的筛选条件更为严格,有效降低了候选轨迹集的规模.针对现有轨迹相似度计算方法效率低下的问题,提出一种基于时差约束的轨迹相似度计算方法.利用轨迹之间的时差排除大量不必要的位置比较运算,将轨迹相似度的计算复杂度控制在线性级别,大大提高了计算效率,同时为过滤伴随轨迹查询过程中的无效计算,对基于时差约束的轨迹相似度计算方法进行变体得到一种上下界过滤方法,最大限度地避免了无效计算.最后,在4个真实的大规模轨迹数据集上对所提方法进行实验验证,实验结果表明所提方法的轨迹查找效率是已知最好方法的9~20倍,证明了算法的有效性.

联合高阶目标感知与相似匹配的目标跟踪算法

视觉目标跟踪算法利用自注意力机制增强上下文联系,但面对复杂场景时,自注意力机制中的相关性易发生失配,为此提出一种联合高阶目标感知与相似匹配的目标跟踪算法。构建高阶目标感知模型,针对自注意力机制中的一阶自注意图,利用坍塌的极化过滤方式进行空间和通道维度的正交化建模,优化内部相关性;同时组合非线性拟合函数避免坍塌引起的信息损失,进而获得高阶自注意图,捕获具有高阶上下文信息的感知特征。通过不同维度分解目标的感知特征来细化匹配区域,抑制背景噪声并约束当前帧的响应图,提高网络的判别力。在OTB100和UAV123基准的实验结果表明,所提算法有更好的跟踪性能,可以有效应对相似干扰等问题。

基于知识注入提示学习的专利短语相似度计算

专利是授予发明者在一定时期内保护其发明的法定权利,在当今的社会活动中发挥着重要作用。然而现有研究并未针对专利相似度数据进行适配优化,导致其应用在专利短语相似度匹配任务中效果不佳。已有研究表明,在低资源的场景下,提示学习将文本片段(模板)作为输入,将分类问题转换为掩码语言建模问题,其关键的一步是在标签空间和标签词空间之间构造一个投影。提出一种基于知识注入的提示学习方法,将其应用于专利短语相似度匹配计算任务。为解决专利短语信息不足的问题,利用专利短语中的相似度标签信息,使用知识增强专利短语与标签信息。首先通过实体链接技术建立专利短语与外部知识的关联关系;然后设计一种基于实体影响度的邻域信息过滤机制,用于缓解专利短语信息不足的问题;最后考虑不同外部知识对专利短语相似度计算的影响,设计应用于专利短语的多种增强提示文本。实验结果表明,该方法的Pеarson相关系数(PCC)和Spеarman相关系数(SRC)相较次优对比方法分别提升6.8%和5.7%。