A methodology for learning across application domains for databasedesign systems

Although database design tools have been developed that attempt to automate (or semiautomate) the design process, these tools do not have the capability to capture common sense knowledge about business applications and store it in a context-specific manner. As a result, they rely on the user to provide a great deal of "trivial" details and do not function as well as a human designer who usually has some general knowledge of how an application might work based on his or her common sense knowledge of the real world. Common sense knowledge could be used by a database design system to validate and improve the quality of an existing design or even generate new designs. This requires that context-specific information about different database design applications be stored and generalized into information about specific application domains (e.g., pharmacy, daycare, hospital, university, manufacturing). Such information should be stored at the appropriate level of generality in a hierarchically structured knowledge base so that it can be inherited by the subdomains below. For this to occur, two types of learning must take place. First, knowledge about a particular application domain that is acquired from specific applications within that domain are generalized into a domain node (e.g., entities, relationships, and attributes from various hospital applications are generalized to a hospital node). This is referred to as within domain learning. Second, the information common to two (or more) related application domain nodes is generalized to a higher-level node; for example, knowledge from the car rental and video rental domains may be generalized to a rental node. This is called across domain learning. This paper presents a methodology for learning across different application domains based on a distance measure. The parameters used in this methodology were refined by testing on a set of representative cases; empirical testing provided further validation     

跨数据集评估的高光谱图像分类

目的 随着高光谱成像技术的飞速发展,高光谱数据的应用越来越广泛,各场景高光谱图像的应用对高精度详细标注的需求也越来越旺盛。现有高光谱分类模型的发展大多集中于有监督学习,大多数方法都在单个高光谱数据立方中进行训练和评估。由于不同高光谱数据采集场景不同且地物类别不一致,已训练好的模型并不能直接迁移至新的数据集得到可靠标注,这也限制了高光谱图像分类模型的进一步发展。本文提出跨数据集对高光谱分类模型进行训练和评估的模式。方法 受零样本学习的启发,本文引入高光谱类别标签的语义信息,拟通过将不同数据集的原始数据及标签信息分别映射至同一特征空间以建立已知类别和未知类别的关联,再通过将训练数据集的两部分特征映射至统一的嵌入空间学习高光谱图像视觉特征和类别标签语义特征的对应关系,即可将该对应关系应用于测试数据集进行标签推理。结果 实验在一对同传感器采集的数据集上完成,比较分析了语义—视觉特征映射和视觉—语义特征映射方向,对比了5种基于零样本学习的特征映射方法,在高光谱图像分类任务中实现了对分类模型在不同数据集上的训练和评估。结论 实验结果表明,本文提出的基于零样本学习的高光谱分类模型可以实现跨数据集对分类模型进行训练和评估,在高光谱图像分类任务中具有一定的发展潜力。     

A new hierarchical packet classification algorithm

Packet classification is one of the most challenging functions in Internet routers since it involves a multi-dimensional search that should be performed at wire-speed. Hierarchical packet classification is an effective solution which reduces the search space significantly whenever a field search is completed. However, the hierarchical approach using binary tries has two intrinsic problems: back-tracking and empty internal nodes. To avoid back-tracking, the hierarchical set-pruning trie applies rule copy, and the grid-of-tries uses pre-computed switch pointers. However, none of the known hierarchical algorithms simultaneously avoids empty internal nodes and back-tracking. This paper describes various packet classification algorithms and proposes a new efficient packet classification algorithm using the hierarchical approach. In the proposed algorithm, a hierarchical binary search tree, which does not involve empty internal nodes, is constructed for the pruned set of rules. Hence, both back-tracking and empty internal nodes are avoided in the proposed algorithm. Two refinement techniques are also proposed; one for reducing the rule copy caused by the set-pruning and the other for avoiding rule copy. Simulation results show that the proposed algorithm provides an improvement in search performance without increasing the memory requirement compared with other existing hierarchical algorithms.     

Multidimensional hierarchical classification

Hierarchical classification can be seen as a multidimensional classification problem where the objective is to predict a class, or set of classes, according to a taxonomy. There have been different proposals for hierarchical classification, including local and global approaches. Local approaches can suffer from the inconsistency problem, that is, if a local classifier has a wrong prediction, the error propagates down the hierarchy. Global approaches tend to produce more complex models. In this paper, we propose an alternative approach inspired in multidimensional classification. It starts by building a multi-class classifier per each parent node in the hierarchy. In the classification phase, all the local classifiers are applied simultaneously to each instance, providing a probability for each class in the taxonomy. Then the probability of the subset of classes, for each path in the hierarchy, is obtained by combining the local classifiers results. The path with highest probability is returned as the result for all the levels in the hierarchy. As an extension of the proposal method, we also developed a new technique, based on information gain, to classifies at different levels in the hierarchy. The proposed method was tested on different hierarchical classification data sets and was compared against state-of-the-art methods, resulting in superior predictive performance and/or efficiency to the other approaches in all the datasets.     

多智能体分层强化学习综述

作为机器学习和人工智能领域的一个重要分支,多智能体分层强化学习以一种通用的形式将多智能体的协作能力与强化学习的决策能力相结合,并通过将复杂的强化学习问题分解成若干个子问题并分别解决,可以有效解决空间维数灾难问题。这也使得多智能体分层强化学习成为解决大规模复杂背景下智能决策问题的一种潜在途径。首先对多智能体分层强化学习中涉及的主要技术进行阐述,包括强化学习、半马尔可夫决策过程和多智能体强化学习;然后基于分层的角度,对基于选项、基于分层抽象机、基于值函数分解和基于端到端等4种多智能体分层强化学习方法的算法原理和研究现状进行了综述;最后介绍了多智能体分层强化学习在机器人控制、博弈决策以及任务规划等领域的应用现状。     

A hierarchical knowledge based system for airplane classification

Airplane classification is used as an application domain to illustrate how hierarchical reasoning on large knowledge bases can be implemented. The knowledge base is organized as a two-dimensional hierarchy: one dimension corresponds to the levels of complexity often seen in computer vision, and the other dimension corresponds to the complexity of hypothesis used in the reasoning process. Reasoning proceeds top-down, from more abstract levels with fewer details toward levels with more details. Whenever possible, with the help of domain knowledge, decision is taken at a higher level, which significantly reduces processing time. A software package called RuBICS (Rule-Based Image Classification System) is described, and some examples of airplane classification are shown     

Flexible syntactic matching of curves and its application toautomatic hierarchical classification of silhouettes

Curve matching is one instance of the fundamental correspondence problem. Our flexible algorithm is designed to match curves under substantial deformations and arbitrary large scaling and rigid transformations. A syntactic representation is constructed for both curves and an edit transformation which maps one curve to the other is found using dynamic programming. We present extensive experiments where we apply the algorithm to silhouette matching. In these experiments, we examine partial occlusion, viewpoint variation, articulation, and class matching (where silhouettes of similar objects are matched). Based on the qualitative syntactic matching, we define a dissimilarity measure and we compute it for every pair of images in a database of 121 images. We use this experiment to objectively evaluate our algorithm. First, we compare our results to those reported by others. Second, we use the dissimilarity values in order to organize the image database into shape categories. The veridical hierarchical organization stands as evidence to the quality of our matching and similarity estimation     

Semantic data integration in hierarchical domains

A major challenge in building the Semantic Web is resolving differences among heterogeneous databases. This article describes one approach for handling semantic data integration problems in hierarchical domains. It also describes a declarative approach for specifying pairwise mappings between a centrally maintained ontology and each local data repository maintained by an autonomous agency. In this context, it outlines a method for specifying the mappings' semantics and encoding them to resolve heterogeneities. It focuses on XML-based applications in which entities in the centrally maintained ontology are hierarchically related to those in the local data repositories.     

Energy-aware QoS for application sessions across multiple protocol domains in mobile computing

The proliferation of heterogeneous devices and diverse networking technologies demands flexible models to guarantee the quality-of-service (QoS) at the application session level, which is a common behavior of many network-centric applications, e.g., Web browsing and Instant messaging. Several QoS models have been proposed for heterogeneous wired/wireless environments. However, we envision that the missing part, which is also a big challenge, is taking energy, a scarce resource for mobile and energy-constrained devices, into consideration. In this paper we propose a novel energy-aware QoS model, e-QoS, for application sessions that might across multiple protocol domains, which will be common in the future Internet, rather than an exception. The model provides QoS guarantee by dynamically selecting and adapting application protocols. To the best of our knowledge, our model is the first attempt to address QoS adaptation at the application session level by introducing a new QoS metric called session lifetime. To show the effectiveness of the proposed scheme, we have implemented two case studies: Web browsing from a Pocket PC to a regular Web server, and an instant messaging application between two Pocket PCs. In the former case study, our approach outperforms the conventional approach without energy-aware QoS by more than 30% in terms of the session lifetime. In the second case study, we also successfully extend the session lifetime to the value negotiated by two Pocket PCs with very diverse battery capacities.     

A supervised learning algorithm for hierarchical classification of fuzzy patterns

The notion of “fuzzy separability” is introduced for fuzzy sets of patterns. A supervised learning algorithm is proposed for estimation of membership functions that yield hierarchical partitioning of the feature space for fuzzy separable pattern classes under confusion. Finally we present a methodology for the design of a classifier composed of hierarchical binary decision trees.     

A comparison of hierarchical multi-output recognition approaches for anuran classification

In bioacoustic recognition approaches, a “flat” classifier is usually trained to recognize several species of anurans, where the number of classes is equal to the number of species. Consequently, the complexity of the classification function increases proportionally with the number of species. To avoid this issue, we propose a “hierarchical” approach that decomposes the problem into three taxonomic levels: the family, the genus, and the species. To accomplish this, we transform the original single-labelled problem into a multi-output problem (multi-label and multi-class) considering the biological taxonomy of the species. We then develop a top-down method using a set of classifiers organized as a hierarchical tree. We test and compare two hierarchical methods, using (1) one classifier per parent node and (2) one classifier per level, against a flat approach. Thus, we conclude that it is possible to predict the same set of species as a flat classifier, and additionally obtain new information about the samples and their taxonomic relationship. This helps us to better understand the problem and achieve additional conclusions by the inspection of the confusion matrices at the three classification levels. In addition, we propose a soft decision rule based on the joint probabilities of hierarchy pathways. With this we are able to identify and reject confusing cases. We carry out our experiments using cross-validation performed by individuals. This form of CV avoids mixing syllables that belong to the same specimens in the testing and training sets, preventing an overestimate of the accuracy and generalizing the predictive capabilities of the system. We tested our methods in a dataset with sixty individual frogs, from ten different species, eight genera, and four families, achieving a final Macro-Fscore of 80 and 70% with and without applying the rejection rule, respectively.     

不平衡分类问题研究综述

实际的分类问题往往都是不平衡分类问题,采用传统的分类方法,难以得到满意的分类效果。为此,十多年来,人们相继提出了各种解决方案。对国内外不平衡分类问题的研究做了比较详细地综述,讨论了数据不平衡性引发的问题,介绍了目前几种主要的解决方案。通过仿真实验,比较了具有代表性的重采样法、代价敏感学习、训练集划分以及分类器集成在3个实际的不平衡数据集上的分类性能,发现训练集划分和分类器集成方法能较好地处理不平衡数据集,给出了针对不平衡分类问题的分类器评测指标和将来的工作。     

A hierarchical human detection system in (un)compressed domains

We propose a hierarchical retrieval system where shape, color and motion characteristics of the human body are captured in compressed and uncompressed domains. The proposed retrieval method provides human detection and activity recognition at different resolution levels from low complexity to low false rates and connects low level features to high level semantics by developing relational object and activity presentations. The available information of standard video compression algorithms are used in order to reduce the amount of time and storage needed for the information retrieval. The principal component analysis is used for activity recognition using MPEG motion vectors and results are presented for walking, kicking, and running to demonstrate that the classification among activities is clearly visible. For low resolution and monochrome images it is demonstrated that the structural information of human silhouettes can be captured from AC-DCT coefficients.     

面向新领域的事件抽取研究综述

在当前互联网时代,大量新领域下的非结构文本数据中蕴含了海量信息.面向新领域的事件抽取方法研究能快速地构建领域知识库,用于支撑基于知识的下游应用.但现有事件抽取系统的领域限定性强,在新领域中从零构建会极度依赖事件体系和标注数据的质量及规模,需要大量人力和专家知识来定制模板和标注语料.而且数据集中常见在相同的上下文中出现多...     

A theory of learning from different domains

Discriminative learning methods for classification perform well when training and test data are drawn from the same distribution. Often, however, we have plentiful labeled training data from a source domain but wish to learn a classifier which performs well on a target domain with a different distribution and little or no labeled training data. In this work we investigate two questions. First, under what conditions can a classifier trained from source data be expected to perform well on target data? Second, given a small amount of labeled target data, how should we combine it during training with the large amount of labeled source data to achieve the lowest target error at test time? We address the first question by bounding a classifier’s target error in terms of its source error and the divergence between the two domains. We give a classifier-induced divergence measure that can be estimated from finite, unlabeled samples from the domains. Under the assumption that there exists some hypothesis that performs well in both domains, we show that this quantity together with the empirical source error characterize the target error of a source-trained classifier. We answer the second question by bounding the target error of a model which minimizes a convex combination of the empirical source and target errors. Previous theoretical work has considered minimizing just the source error, just the target error, or weighting instances from the two domains equally. We show how to choose the optimal combination of source and target error as a function of the divergence, the sample sizes of both domains, and the complexity of the hypothesis class. The resulting bound generalizes the previously studied cases and is always at least as tight as a bound which considers minimizing only the target error or an equal weighting of source and target errors.     

Decision trees for hierarchical multi-label classification

Hierarchical multi-label classification (HMC) is a variant of classification where instances may belong to multiple classes at the same time and these classes are organized in a hierarchy. This article presents several approaches to the induction of decision trees for HMC, as well as an empirical study of their use in functional genomics. We compare learning a single HMC tree (which makes predictions for all classes together) to two approaches that learn a set of regular classification trees (one for each class). The first approach defines an independent single-label classification task for each class (SC). Obviously, the hierarchy introduces dependencies between the classes. While they are ignored by the first approach, they are exploited by the second approach, named hierarchical single-label classification (HSC). Depending on the application at hand, the hierarchy of classes can be such that each class has at most one parent (tree structure) or such that classes may have multiple parents (DAG structure). The latter case has not been considered before and we show how the HMC and HSC approaches can be modified to support this setting. We compare the three approaches on 24 yeast data sets using as classification schemes MIPS’s FunCat (tree structure) and the Gene Ontology (DAG structure). We show that HMC trees outperform HSC and SC trees along three dimensions: predictive accuracy, model size, and induction time. We conclude that HMC trees should definitely be considered in HMC tasks where interpretable models are desired.     

Fingerprint classification by a hierarchical classifier

Fingerprint classification is still a challenging problem due to large intra-class variability, small inter-class variability and the presence of noise. To deal with these difficulties, we propose a regularized orientation diffusion model for fingerprint orientation extraction and a hierarchical classifier for fingerprint classification in this paper. The proposed classification algorithm is composed of five cascading stages. The first stage rapidly distinguishes a majority of Arch by using complex filter responses. The second stage distinguishes a majority of Whorl by using core points and ridge line flow classifier. In the third stage, K-NN classifier finds the top two categories by using orientation field and complex filter responses. In the fourth stage, ridge line flow classifier is used to distinguish Loop from other classes except Whorl. SVM is adopted to make the final classification in the last stage. The regularized orientation diffusion model has been evaluated on a web-based automated evaluation system FVC-onGoing, and a promising result is obtained. The classification method has been evaluated on the NIST SD 4. It achieved a classification accuracy of 95.9% for five-class classification and 97.2% for four-class classification without rejection.     

Computer-assisted pit-pattern classification in different wavelet domains for supporting dignity assessment of colonic polyps

In this paper, we show that zoom-endoscopy images can be well classified according to the pit-pattern classification scheme by using texture-analysis methods in different wavelet domains. We base our approach on three different variants of the wavelet transform and propose that the color channels of the RGB and LAB color model are an important source for computing image features with high discriminative power. Color-channel information is incorporated by either using simple feature vector concatenation and cross-cooccurrence matrices in the wavelet domain. Our experimental results based on k-nearest neighbor classification and forward feature selection exemplify the advantages of the different wavelet transforms and show that color-image analysis is superior to grayscale-image analysis regarding our medical image classification problem.     

单路径层次化网页分类算法

网页自动分类是解决互联网信息检索困难的有效方法.虽然有很多自动分类算法和系统,但是大部分此类算法注重如何将网页准确分到某个独立的类别里面,却忽略类别之间所组成的体系结构本身也具备的一些隐藏分类信息.同时,一般的分类算法每次分类都需要搜索所有的类别.针对这些缺点,提出了一种基于结构的单路径层次化网页分类算法,该分类方法利用类别之间具有树状结构这一特点,对类别中存在父子关系的类别间进行信息传递,使得每次分类只需要搜索树中一条路径而不用遍历所有树节点.实验结果证明,这种单路径搜索技术与相关的算法相比,在减少搜索节点的同时可以提高6%的准确度.