Fuzzy knowledge representation study for incremental learning in data streams and classification problems

The extraction of models from data streams has become a hot topic in data mining due to the proliferation of problems in which data are made available online. This has led to the design of several systems that create data models online. A novel approach to online learning of data streams can be found in Fuzzy-UCS, a young Michigan-style fuzzy-classifier system that has recently demonstrated to be highly competitive in extracting classification models from complex domains. Despite the promising results reported for Fuzzy-UCS, there still remain some hot issues that need to be analyzed in detail. This paper carefully studies two key aspects in Fuzzy-UCS: the ability of the system to learn models from data streams where concepts change over time and the behavior of different fuzzy representations. Four fuzzy representations that move through the dimensions of flexibility and interpretability are included in the system. The behavior of the different representations on a problem with concept changes is studied and compared to other machine learning techniques prepared to deal with these types of problems. Thereafter, the comparison is extended to a large collection of real-world problems, and a close examination of which problem characteristics benefit or affect the different representations is conducted. The overall results show that Fuzzy-UCS can effectively deal with problems with concept changes and lead to different interesting conclusions on the particular behavior of each representation.     

A knowledge representation for constraint satisfaction problems

We present a general representation for problems that can be reduced to constraint satisfaction problems (CSP) and a model for reasoning about their solution. The novel part of the model is a constraint-driven reasoner that manages a set of constraints specified in terms of arbitrarily complex Boolean expressions and represented in the form of a dependency network. This dependency network incorporates control information (derived from the syntax of the constraints) that is used for constraint propagation, contains dependency information that can be used for explanation and for dependency-directed backtracking, and is incremental in the sense that if the problem specification is modified, a new solution can be derived by modifying the existing solution. The constraint-driven reasoner is coupled to a problem solver which contains information about the problem variables and preference orderings     

基于自适应权重的多重稀疏表示分类算法

提出了一种基于多特征字典的稀疏表示算法。该算法针对SRC的单特征鉴别性较弱这一不足,对样本提出多个不同特征并分别进行相应的稀疏表示。并根据SRC算法计算各个特征的鉴别性,自适应地学习出稀疏权重并进行线性加权,从而提高分类的性能。实验表明,基于自适应权重的多重稀疏表示分类算法,具有更好的分类效果。     

基于知识库分割的多知识库整合方法

知识库是集团企业云制造平台中知识服务的重要基础,知识库的质量直接决定着知识服务的质量。目前单一分散的知识库不能提供统一的知识资源全局视图,不利于知识资源的共享。多知识库整合技术已经成为该领域的研究热点之一。提出一种基于知识库分割的多知识库整合方法,采用基于蚁群聚类的分割策略,将知识库有效划分为知识块集。在知识块间利用语义概念映射生成知识块间映射图,从而实现多知识库整合。通过对算法的时间复杂度进行分析,表明该方法在时间复杂度方面要优于基于最小概念集的多知识库整合方法;实验结果也表明该方法在运行性能方面明显优于已有的方法。     

Tackling representation,annotation and classification challenges for temporal knowledge base population

Temporal Information Extraction (TIE) plays an important role in many natural language processing and database applications. Temporal slot filling (TSF) is a new and ambitious TIE task prepared for the knowledge base population (KBP2011) track of NIST Text Analysis Conference. TSF requires systems to discover temporally bound facts about entities and their attributes in order to populate a structured knowledge base. In this paper, we will provide an overview of the unique challenges of this new task and our novel approaches to address these challenges. We present challenges from three perspectives: (1) Temporal information representation: We will review the relevant linguistic semantic theories of temporal information and their limitations, motivating the need to develop a new (4-tuple) representation framework for the task. (2) Annotation acquisition: The lack of substantial labeled training data for supervised learning is a limiting factor in the design of TSF systems. Our work examines the use of multi-class logistic regression methods to improve the labeling quality of training data obtained by distant supervision. (3) Temporal information classification: Another key challenge lies in capturing relations between salient text elements separated by a long context. We develop two approaches for temporal classification and combine them through cross-document aggregation: a flat approach that uses lexical context and shallow dependency features and a structured approach that captures long syntactic contexts by using a dependency path kernel tailored for this task. Experimental results demonstrated that our annotation enhancement approach dramatically increased the speed of the training procedure (by almost 100 times), and that the flat and structured classification approaches were complementary, together yielding a state-of-the-art TSF system.     

Class-specific image representation for image classification using multiple scale-invariant region detectors

We propose a new class-specific image representation for image classification using multiple region detectors. The new representation is designed to solve the problem of increasing variation in object location and size within images of a class, for which traditional spatial pyramid matching shows limited classification accuracy. We propose a new region-division method that divides the image region into two class-specific regions, called class-specific region-of-interest (C-ROI) and focal region (FR). Using multiple region detectors and appropriate mixing of their responses avoids the problem of selecting a region detector that gives the best classification accuracy for a given image class, and thereby yields better results than using only one region detector. Several scale-invariant region detectors are used to obtain C-ROI and FR by considering their importance over a given image class. In experiments using several well-known datasets, the proposed method improved the accuracy and achieved results that were better than or comparable to those achieved by the related methods.     

Propositional representation for graphical knowledge

Multi-media interfaces with a graphics and a natural language component can be viewed as a Natural Language Graphics systems without a host program. We will investigate a theory of Natural Language Graphics that is based on the notion of “Graphical Deep Knowledge” defined in this research. Graphical Deep Knowledge is knowledge that can be used for display purposes as well as reasoning purposes and we describe the syntax and semantics of its constructs. This analysis covers forms, positions, attributes, parts, classes, reference frames, inheritability, etc. Part hierarchies are differentiated into three sub-types. The usefulness of inheritance along part hierarchies is demonstrated, and criticism of inheritance-based knowledge representation formalisms with a bias towards class hierarchies is derived from this finding. The presented theory has been implemented as a generator program that creates pictures from knowledge structures, and as an augmented transition network grammar that creates knowledge structures from limited natural language input. The function of the picture generation program Tina as a user interface for a circuit board maintenance system and as part of a CAD-like layout system is demonstrated.     

基于稀疏近邻表示的分类方法

稀疏表示分类方法(SRC)在人脸识别方面取得了当前最好的分类结果,针对SRC存在的问题,提出稀疏近邻表示方法(SNRC).在局部线性嵌入方法前提假设成立的条件下,SNRC通过稀疏近邻表示实现目标分类.在几个不同数据集上的实验结果显示,SNRC适用于呈非线性分布的数据集,并取得了较好的效果.进一步的分析表明,SNRC能够较好的适用于那些通过降维方法得到的低维数据的分类问题,尤其适用于基于近邻保持的一类降维方法得到的低维数据,并且具有较低的时间复杂度.     

Sparse neighbor representation for classification

Recent research of sparse signal representation has aimed at learning discriminative sparse models instead of purely reconstructive ones for classification tasks, such as sparse representation based classification (SRC) which obtains state-of-the-art results in face recognition. In this paper, a new method is proposed in that direction. With the assumption of locally linear embedding, the proposed method achieves the classification goal via sparse neighbor representation, combining the reconstruction property, sparsity and discrimination power. The experiments on several data sets are performed and results show that the proposed method is acceptable for nonlinear data sets. Further, it is argued that the proposed method is well suited for the classification of low dimensional data dimensionally reduced by dimensionality reduction methods, especially the methods obtaining the low dimensional and neighborhood preserving embeddings, and it costs less time.     

融合局部思想和协作表达的鲁棒分类

目的 协作表达分类算法在人脸识别实验上表现出较好的性能,但其未考虑样本的局部特性,且算法只能处理测试样本中的噪声,未能有效处理训练样本集中的噪声.针对这两个问题,提出融合局部思想和协作表达的鲁棒分类算法.方法 一方面,在训练集上,通过奇异值分解SVD得到其有效表达,丢弃一些噪声;另一方面,算法考虑数据的局部相似性,以保持测试样本与其相邻训练样本之间的相似性.结果 本文算法能得到一个闭式(closed-form),可避免稀疏表示分类算法中由于迭代引起的高时间复杂度问题,在ORL、扩展YALEB和PIE人脸库上的识别率分别可达91.4%,93.8%和93.2%,与同类算法相比识别率有较大幅度地提高;实验结果验证了算法所得到的系数具有较高的判别能力.结论 算法将训练样本进行奇异值分解得到“干净”的训练样本,能在一定程度上消除噪声的影响,且在协作表达的基础上,考虑测试样本和与之相邻的训练样本的局部相似性,相比原始的协作表达分类算法有更好的稳定性和鲁棒性.     

Integration of prior knowledge of measurement noise in kernel density classification

Samples can be measured with different precisions and reliabilities in different experiments, or even within the same experiment. These varying levels of measurement noise may deteriorate the performance of a pattern recognition system, if not treated with care. Here we seek to investigate the benefit of incorporating prior knowledge about measurement noise into system construction. We propose a kernel density classifier which integrates such prior knowledge. Instead of using an identical kernel for each sample, we transform the prior knowledge into a distinct kernel for each sample. The integration procedure is straightforward and easy to interpret. In addition, we show how to estimate the diverse measurement noise levels in a real world dataset. Compared to the basic methods, the new kernel density classifier can give a significantly better classification performance. As expected, this improvement is more obvious for small sample size datasets and large number of features.     

Edge-preserving-based collaborative representation for spectral-spatial classification

Recently, the nearest regularized subspace (NRS) classifier and its spectral–spatial versions such as joint collaborative representation (JCR) and weighted JCR (WJCR) have gained an increasing interest in the hyperspectral image classification. JCR and WJCR average each pixel with its neighbours in a spatial neighbourhood window. The use of spatial information as averaging of pixels in a local window may degrade the classification accuracy in the neighbourhood of discontinuities and class boundaries. We propose the edge-preserving-based collaborative representation (EPCR) classifier in this article, which overcomes this problem by using the edge image estimated by the original full-band hyperspectral image. The estimated edge image is used for calculation of the weights of neighbours and also the final residuals in the collaborative representation classifier. The advantage of multiscale spatial window is also assessed in this work. Moreover, the kernelized versions of NRS and its improved versions are developed in this article. Our experimental results on several popular hyperspectral images indicate that EPCR and its kernelized version are superior to some state-of-the-art classification methods.     

A bag-of-regions representation for video classification

A bag-of-regions (BoR) representation of a video sequence is a spatio-temporal tessellation for use in high-level applications such as video classifications and action recognitions. We obtain a BoR representation of a video sequence by extracting regions that exist in the majority of its frames and largely correspond to a single object. First, the significant regions are obtained using unsupervised frame segmentation based on the JSEG method. A tracking algorithm for splitting and merging the regions is then used to generate a relational graph of all regions in the segmented sequence. Finally, we perform a connectivity analysis on this graph to select the most significant regions, which are then used to create a high-level representation of the video sequence. We evaluated our representation using a SVM classifier for the video classification and achieved about 85 % average precision using the UCF50 dataset.     

一种基于联合表示的图像分类方法

在图像分类识别中,对于同一目标的不同图像,其训练样本和测试样本在同一位置的像素强度通常不同,这不利于提取目标图像的显著特征。这里给出一种基于稀疏表示的联合表示的图像分类方法,此方法首先利用相邻列之间的关系得到原始图像对应的虚拟图像,利用虚拟图像提高图像中中等强度像素的作用,降低过大或过小强度像素对图像分类的影响;然后用同一个目标的原始图像和虚拟图像一起表示目标,得到目标图像的联合表示;最后利用联合表示方法对目标分类。针对不同目标图像库的实验研究表明,给出的联合方法优于利用单一图像进行分类的方法,而且本方法能联合不同的表示方法来提高图像分类正确率。     

The evidence algorithm and problems of representation and processing of mathematical computer knowledge

A series of research works on the realization of the “evidence algorithm” program initiated by V. M. Glushkov in the sixties is described. The works are planned in the context of a new understanding of the theorem-proving problem and according to modern trends in information technologies and computer science. Translated from Kibemetika i Sistemnyi Analiz, No. 6, pp. 9–17, November–December, 1999.     

Towards a cognitive foundation for knowledge representation

Abstract. Knowledge engineering, knowledge management and conceptual modelling are concerned with representing knowledge of business and organizational domains. These research areas use ontologies for knowledge representation. Ontologies are understood either in the philosophical sense as firm metaphysical commitments or in the looser sense of dictionaries or taxonomies.
This paper critically examines the understanding and use of ontologies and knowledge representation languages in information systems (IS) research and application. As ontologies are intended to be conceptualizations of a perceived reality, they should reflect the empirically observed reality. This motivates proposing psychology of language as a reference discipline for knowledge engineering and knowledge management. Natural language is argued to reflect the cognitive concepts we use to think about and perceive the world around us. These cognitive concepts are the relevant terms with which to structure and represent knowledge about the world.
Psychology of language can provide empirical justification for a particular set of concepts to represent knowledge. This paper draws on psycho-linguistic research to develop a proposal for a system of cognitive structures. This is argued to provide the relevant concepts on which to found knowledge representation schemata for knowledge engineering, knowledge management and conceptual modelling.     

A decidable first-order logic for knowledge representation

Decidable first-order logics with reasonable model-theoretic semantics have several benefits for knowledge representation. These logics have the expressive power of standard first order logic along with an inference algorithm that will always terminate, both important considerations for knowledge representation. Knowledge representation systems that include a faithful implementation of one of these logics can also use its model-theoretic semantics to provide meanings for the data they store. One such logic, a variant of a simple type of first-order relevance logic, is developed and its properties described. This logic, although extremely weak, does capture a non-trivial and well-motivated set of inferences that can be entrusted to a knowledge representation system.This is a revised and much extended version of a paper of the same name that appears in the Proceedings of the Ninth International Joint Conference on Artificial Intelligence, Los Angeles, California, 1985.     

面向对象的专家知识模型

文中用面向对象方法来分析、设计基于规则的专家系统,提出一种新的采用分层递阶结构的专家知识模型,并从结构和功能角度进行详细阐述。