关系分类是自然语言处理中一项重要的基础性任务,目的是识别实体对之间的语义关系.目前的方法主要依赖于句子特征,忽视了句子中实体的信息,而句子中的实体位置信息、实体类型信息以及实体依存信息等多元实体信息有助于识别实体间关系.为充分利用实体信息,提出一种融入多元实体信息关系分类模型BERT-MEI.首先标记实体类型和提取实体最短依存路径,然后通过预训练的语言表征(Bidirectional Encoder Representation from Transform-ers,BERT)模型编码,将编码后的句子向量、实体向量和实体依存关系向量合并为最终的实体关系表示.在KBP37数据集和TACRED数据集上的实验结果表明,BERT-MEI模型的F1值比基线模型提高了1～17百分点,验证了利用多元实体信息,能够提升关系分类的效果.

Human brain function parcellation is an important way to reveal the separation of brain functions. However, most of the existing parcellation methods can not deal with the high dimension and low signal-to-noise ratio of functional magnetic resonance imaging (fMRI) data, so they show the problem of weaker search ability and poorer parcellation results. To alleviate this problem, a human brain functional parcellation method based on artificial jellyfish search optimization (AJSO) algorithm is proposed. Firstly, a functional correlation matrix is calculated based on the preprocessed fMRI data and mapped to a low-dimensional space. Then, a food is encoded as a cluster solution composed of multiple functional cluster centers and the improved AJSO is used to search for better food. The time control mechanism integrated with iterative stagnation is used to control the artificial jellyfish to perform active or passive motion, so as to improve the global search ability. Step size determination strategy guided by fitness is designed for active movement to enhance scientific and targeted search of artificial jellyfish. Finally, according to the principle of minimum distance, the cluster label of each row data in the correlation matrix is obtained and mapped to the corresponding voxels. Experiments on real fMRI data show that compared with other partitioning methods, the new method not only has higher searching ability, but also can obtain better spatial structures and stronger functional consistency. In this study, artificial jellyfish search optimization algorithm is applied to brain functional parcellation, which provides a more effective method of brain functional parcellation.

With the development of technologies such as Internet of things, cloud computing and artificial intelligence, big data becomes a research hotspot and is applied in many fields. The booming aviation field has natural big data soil, which has been paid more and more attention. In recent years, scholars have begun to study aviation oriented big data technology. Meanwhile airlines have also begun to use aviation big data to provide services for them, and promote it to the height of development strategy. Research and practice have shown that aviation big data can not only help to reduce the company??s operating costs, but also improve the quality of customer experience. In this paper, the definition of aviation big data is firstly given from the perspective of data and system, and the corresponding organization structures are described systematically. Secondly, the key technologies of aviation big data are elaborated in detail from five aspects: collection, storage management, preprocessing, analysis and virtual simulation and visualization, and some main models and algorithms are compared and analyzed. This paper describes the typical application scenarios of aviation big data from many aspects. Finally, the problems existing in aviation big data and the future research directions are analyzed in an in-depth way, so as to provide useful references for related research and applications.

Face recognition has many applications in pattern recognition and computer vision, and many face recognition methods have been proposed. Among them, the recently proposed collaborative representation based face recognition has attracted the attention of researchers. Many variants and extensions of collaborative representation based classification (CRC) have been presented. However, most of CRC methods do not consider data locality, which is crucial for classification task. In this article, a novel collaborative representation based face recognition method, LP-CRC, is proposed, which balances data locality and collaborative representation. The proposed method incorporates a locality adaptor term into the robust collaborative representation based classification framework, leading to a novel unified objective function. The Augmented Lagrange Multiplier is used to optimize the objective function. Tests on standard benchmarks demonstrate that the proposed face recognition method is superior to existing methods and robust to noise and outliers.

培养双创型人才是建设创新型国家的客观要求,人才培养模式改革是高素质人才培养的必然趋势.概述了双创型人才和人才培养模式的内涵,剖析了新型计算机人才培养面临的困境.提出了一种新的双创型计算机人才培养模式,从制定多层次培养方案、建立立体化实践教学体系、优化教育资源、构建多元化人才评价机制和搭建合理的双创教育平台五大方面分析了该模式的运行方式.教学实践表明,这些措施有利于双创人才素质的提高.

In this study, a whitening transformation based approach to one-class classification of remote sensing imagery is investigated. Only positive data are required to train the one-class classifier. Firstly, the image data is mapped to a new feature space using the whitening processing with all directions of the class of interest. Then a threshold is selected to make a binary prediction. A heuristic method of threshold selection is performed in the experiment of one-class classification. A series of values are set to the threshold based on standard deviation, and perform the one-class classification with each threshold value. The experiment shows that high accuracy is achieved with the threshold range from 3 to 4 standard deviations of the mean. Finally, the results of one-class classification with the threshold of 3 standard deviations are compared to that of one-class support vector machine. The results indicate that the proposed method provides nearly the same accuracy of one-class classification as one-class support vector machine. The advantage of the proposed method is that it can use a constant threshold to extract various land types.

Similarity theory based regional segmentation image fusion technique has achieved great success in heterogeneous sensor image fusion.Traditional similarity parameter can not fit the structure information well.This paper induced wavelet similarity to substitute traditional correlation coefficient similarity.Wavelet similarity reflects the detail structure information well in source images.It adopted contrast sensitivity function reflecting the human vision to weight the wavelet coefficient on different scales,divided redundant and complementary regions by weighted similarity.Experimental results demonstrate that the proposed method measures the structural similarity well and obtains better result than traditional similarity based method.

针对CPS的物联性、互联性和智联性问题,提出了具有实时性、灵活性及适应性特点的智能代理CPS数据处理模型.模型从网络系统数据处理的分布式、智能化角度对CPS计算和控制构件进行描述及设计,给出了在数据计算和控制过程中CPS数据处理关键步骤的设计方案.通过对智能电网系统数据的分析,探讨了CPS在智能电网中的应用,以期对CPS实用化提供支持.

先秦时期是陶、玉、青铜三大礼器的形成、发展和辉煌时期。陶、玉礼器均由新石器时代中晚期平等农耕时代的祭器转化而来，其在新石器时代晚期的中心聚落时期登上礼器舞台，并呈现对峙趋势，表现出各有主导的中心区域：黄河流域以彩陶礼器为主导，反映了祖先崇拜的意识形态发展历程；南方长江流域和北方辽河流域以玉礼器为主导，呈现出神灵崇拜的神秘和威严。到新石器时代末期早期国家的龙山时代，礼器在北方以黄河流域为中心的地区表现为陶、玉礼器的融合并重，南方依然以玉礼器为主导。夏商周王国建立以后，随着权力的集中与统一，代表新兴王权特质的青铜礼器迅速崛起和普及，占据了礼器的主导地位，成为国之重器，陶、玉礼器成为附属礼器。

<span id="ChDivSummary" name="ChDivSummary" class="abstract-text">基于传统Web应用架构的研究,结合开源框架的特点和优势,设计了ESSH多层Web应用框架。介绍了猪肉生产的业务流程,阐述了猪肉质量追溯系统的系统结构和功能模块,在此基础上利用ESSH框架开发实现了系统,论文对系统中用户授权模块的开发过程和实现细节进行了详细说明。实践证明,利用这种架构模型开发追溯系统能极大程度上提高系统的稳定性、可移植性和可重用性,从而解决了传统系统架构下的缺陷,具有较好的应用与推广价值。</span>

A scene classification algorithm based on global optimized framework is proposed. Firstly, the global scene feature named spatial envelop is obtained from the whole image, the visual word of each image block is extracted, and latent variable is defined to represent the semantic feature of the extracted visual word. Secondly, the structure graph of latent state is introduced to represent the context of visual words. In respect to scene classification strategy, objective function consisting of different potential functions is constructed in which potential functions are defined to measure the relevance of the variables including global scene feature, latent variables and scene category. Finally, the scene category of the image is determined when the global optimized solution of objective function is obtained. The experiments on the standard dataset demonstrate that the proposed algorithm achieves better results than the state-of-the-art algorithms.

工程教育专业认证有助于优化工程专业人才培养体系,促进专业教学质量的提升.地方高校的计算机专业以工程教育专业认证为契机,通过明确专业定位,制定合理培养方案,构建具有特色的课程体系,丰富教师工程实践经验和提升其专业素质,加强大学生实践能力培养和完善教学质量监控体系,可以全面提升人才培养质量,提升专业竞争力.

A multiple watermarking algorithm is proposed to protect relational databases copyright for some lacks of the existing multi -watermarking algorithms .The novel multiple watermarking scheme ,which embeds multi -media watermarks into relational database .The watermark embedding and extraction algorithms are specified .The algorithm is “blind” in that it isn’ t requires original data in order to detect a watermark . The results of the corresponding watermark experiments and the attack experiments verify that the proposed method is correct feasible and robust .

A new approach to scene classification is proposed based on integrated scale-invariant feature transform(SIFT) feature and visual dictionary using twice-clustering method.Firstly,the proposed integrated SIFT feature operator adds some pseudo-extreme points and non-extreme points to points of interest based on traditional SIFT method,and it can use the more feature points and make the feature points distribution more uniform in the image;Secondly,the features in every image are clustered before the visual dictionary is constructed,and it can make the visual word represents the more scene information and greatly reduce the time of constructing visual dictionary.Finally,the probabilistic latent semantic analysis(PLSA) model is used for training and testing.The test on the standard image dataset shows that the proposed approach has the better classification results,and deal with the different scene categories very well.

The thesis firstly summarizes the principle and realization of Ant-Miner algorithm.And then analyses the Ant-Miner algorithm from different angles and proposed an improving and optimizing method in order to overcome the problems existed in the algorithm.Finally,the experiments show that optimization algorithm can achieve better results.

A comparative study was made on the parameters of the local characteristics of image clarity,and the optimal parameter of the Energy Of Laplacian(EOL) was obtained.A new intelligent image fusion algorithm based on EOL was proposed.A set of registered images was firstly segmented,and then local EOLs of segmented image blocks were computed.EOLs were input into neural network and the target vectors were automatically obtained by comparing values of EOLs.Test images were segmented and their EOLs were put into trained network and the rough fusion images were generated.Final fusion results would be obtained by consistence verification.The experimental results demonstrate the good fusion performance on different source images.

高斯混合模型(GMMs)是统计学习理论的基本模型,在可视媒体领域应用广泛.近些年来,随着可视媒体信息的增长和分析技术的深入,GMMs在(纹理)图像分割、视频分析、图像配准、聚类等领域有了进一步的发展.从GMMs的基本模型出发,从理论和应用的角度讨论和分析了GMMs的求解算法,包括EM算法、变化形式等,论述了GMMs的模型选择问题:在线学习和模型约简.在视觉应用领域,介绍了GMMs在图像分段、视频分析、图像配准、图像降噪等领域的扩展模型与方法,详细地阐述了一些最新的典型模型的原理与过程,如用于图像分段的空间约束GMMs、图像配准中的关联点漂移算法.最后,讨论了一些潜在的发展方向与存在的困难问题.

This paper proposed novel chaos-artificial bee colony algorithm for continuous function optimization problems.Based on the Memetic algorithm framework,the new algorithm took artificial bee colony algorithm as the global search algorithm and the chaos local search operator as the local search operator algorithm.Furthermore,for the food source which was into local minimum,the scout didn't randomly generate the new food source,but used the chaos local operator to generate the candidate food source in order to enhance the exploit of bee colony.The simulation results for five Benchmark functions show that,compared with those of artificial bee colony algorithm,the new algorithm has the advantages to the solutions' quality.

The principle and realization of Ant-Miner algorithm are summarized firstly.Then the Ant-Miner algorithm is analyzed from different views,and an improving and optimizing method are proposed in order to overcome the problems existed in the algorithm.Finally,the improved Ant-Miner algorithm is used in earthquake prediction.The experiments show that,optimization algorithm can achieve better results than C4.5 algorithm.