Class Of Pairs Research Articles

Towards an autonomous human chromosome classification system using Competitive Support Vector Machines Teams (CSVMT)

Abstract In broad terms, karyotyping is the process of examination and classification of human chromosome images to diagnose genetic diseases and disorders. It requires time consuming manual examination of cell images by a cytogeneticist to distinguish chromosome classes from each other. Thus, a reliable autonomous human chromosome classification system not only saves time and money but also reduces errors due to the inadequate knowledge level of the expert. Human cell contains 23 pairs of chromosome, 22 autosomes and a pair of sex chromosomes. Hence, we face a multi-class classification task which represents a challenging case for any sort of classifier. In this work, to solve this classification problem, we propose a novel methodology consisting two stages: (i) data preparation and training, and (ii) testing. To determine the most informative content of the dataset several preliminary experiments are conducted and a Principal Component Analysis is done. Then, a single Support Vector Machine (SVM ij ) is trained to separate a pair of classes, (i,j) where a numerical optimization method Pattern Search (PS), is employed to find the optimal parameters for the SVM ij . Considering 22 pairs of autosomes, 22 × 22 experts are trained and optimized. The cluster of experts, we obtain is named as Competitive SVM Teams (CSVMTs) where each SVM ij competes with the others to label a new classification instance. The final output of the classifier is determined by majority voteing. The results obtained on Copenhagen dataset proves the merit of the algorithm as correct classification rates (CRR) on train and test samples are 99.55% and 97.84% respectively, which are higher than any accuracy rate achieved so far in the related literature.

Are children participating in a quasi-experimental education outside the classroom intervention more physically active?

BackgroundEducation outside the classroom (EOtC) is a curriculum-based approach to teaching that has shown positive associations with children’s physical activity and academic learning in small-scale case studies. The purpose of this large-scale quasi-experimental study was to determine if children who participate regularly in EOtC spend more time being physically active than children who do not.MethodsIn the 2014/2015 study TEACHOUT, classes were recruited in pairs such that each EOtC class had a non-EOtC comparison class at the same school and grade level. Participants in 17 EOtC classes and 16 comparison parallel classes across Denmark wore an Axivity AX3 accelerometer taped to the lower back for seven consecutive days. Data from 201 EOtC participants (63.3% girls, age 10.82 ± 1.05,) and 160 comparison participants (59.3% girls, age 10.95 ± 1.01) were analysed using an ‘intention to treat’ (ITT) approach. The amount of EOtC the participants were exposed to was monitored. Associations between time spent in different physical activity intensities and EOtC group and sex were assessed using generalised linear models adjusted for age. In a second analysis, we modified the sample using a ‘per protocol’ (PP) approach, only including EOtC and comparison class pairs where the EOtC class had >150 min and the comparison had <150 min of EOtC during the measured week.ResultsOn average, EOtC participants spent 8.4 (ITT) and 9.2 (PP) minutes more in moderate-to-vigorous physical activity (MVPA) per day than comparison participants (p < 0.05). However, EOtC boys spent 18.7 (ITT) and 20.8 (PP) minutes more in MVPA per day than comparison boys (p < 0.01), while there were no significant between-group differences for girls.ConclusionsFor boys, EOtC was associated with more daily time being spent moderately and vigorously physically active. No differences were observed for girls. Implementing EOtC into schools’ weekly practice can be a time- and cost-neutral, supplementary way to increase time spent in PA for boys through grades three to six.Trial registrationThe Scientific Ethical Committee in the Capital Region of Denmark protocol number H-4-2014-FSP. 5 March, 2014.

A Ternary Brain-Computer Interface Based on Single-Trial Readiness Potentials of Self-initiated Fine Movements: A Diversified Classification Scheme.

In recent years, the readiness potential (RP), a type of pre-movement neural activity, has been investigated for asynchronous electroencephalogram (EEG)-based brain-computer interfaces (BCIs). Since the RP is attenuated for involuntary movements, a BCI driven by RP alone could facilitate intentional control amid a plethora of unintentional movements. Previous studies have mainly attempted binary single-trial classification of RP. An RP-based BCI with three or more states would expand the options for functional control. Here, we propose a ternary BCI based on single-trial RPs. This BCI classifies amongst an idle state, a left hand and a right hand self-initiated fine movement. A pipeline of spatio-temporal filtering with per participant parameter optimization was used for feature extraction. The ternary classification was decomposed into binary classifications using a decision-directed acyclic graph (DDAG). For each class pair in the DDAG structure, an ordered diversified classifier system (ODCS-DDAG) was used to select the best among various classification algorithms or to combine the results of different classification algorithms. Using EEG data from 14 participants performing self-initiated left or right key presses, punctuated with rest periods, we compared the performance of ODCS-DDAG to a ternary classifier and four popular multiclass decomposition methods using only a single classification algorithm. ODCS-DDAG had the highest performance (0.769 Cohen's Kappa score) and was significantly better than the ternary classifier and two of the four multiclass decomposition methods. Our work supports further study of RP-based BCI for intuitive asynchronous environmental control or augmentative communication.

Decision problems, complexity, traces, and representations

In this article, we study connections between representation theory and efficient solutions to the conjugacy problem on finitely generated groups. The main focus is on the conjugacy problem in conjugacy separable groups, where we measure efficiency in terms of the size of the quotients required to distinguish a distinct pair of conjugacy classes.

Spectral features for audio based vehicle and engine classification

In this paper we address automatic vehicle and engine identification based on audio information. Such data depend on many factors, including vehicle type, tires, speed and its change, as well as road type. In our previous research we designed a feature set for selected vehicle classes, discriminating pairs of classes. Later, we decided to expand the feature vector and find the best feature set (mainly based on spectral descriptors), possibly representative for each investigated vehicle category, which can be applied to a bigger data set, with more classes. The experiments were performed first on on-road recordings, and then continued with test bench (dyno) recordings. The paper also shows problems related to vehicles classification, which is detailed in official documents by national authority for issues related to the national road system, but simplified for automatic identification purposes. Experiments on audio-based vehicle type and engine type identification are presented and conclusions are shown.

Shape indexes for semi-automated detection of windbreaks in thematic tree cover maps from the central United States

Windbreaks are an important ecological resource across the large expanse of agricultural land in the central United States and are often planted in straight-line or L-shaped configurations to serve specific functions. As high-resolution (i.e., <5m) land cover datasets become more available for these areas, semi- or fully-automated methods for distinguishing windbreaks from other patches of trees are needed for use with thematic raster datasets. To address this need, we created three shape indexes: a morphology-based index that we have named the Straight and Narrow Feature Index (SNFI), a windbreak sinuosity index, and an area index indicating the occupied fractional area of a bounding box. The indexes were tested in two study areas: (1) a riparian area dominated by sinuous bands of trees but mixed with row crop agriculture and (2) an agricultural area with a mix of straight-line and L-shaped windbreaks. In the riparian area, a Kruskall–Wallis rank sum test indicated class differences for all three indexes, and pairwise comparisons indicate windbreaks and riparian trees are separable using any of the three indexes. SNFI also produced significant differences between windbreaks oriented in different directions (east–west vs. north–south). In the agricultural area, the Kruskall–Wallis rank sum test indicated differences between classes for all three indexes, and pairwise comparisons show that all class pairs have significant differences for at least one index, with the exception of L-shaped windbreaks vs. non-windbreak tree patches. We also used classification trees to objectively assign representative samples of tree patches to classes using both single indexes and multiple indexes. Classes were correctly assigned for more than 90% of the samples in both the riparian and agricultural study areas. In the riparian area, combining indexes did not improve accuracy compared to using SNFI alone, whereas in the agricultural area, combining the three indexes produced the best result. Thematic datasets derived from high-resolution imagery are becoming more available, and extracting useful information can be a challenge, partly due to the large amount of data to assess. Calculating the three shape indexes presented can assist with efficient identification of candidate windbreaks and as such, hold good promise for value-added analysis of tree function in the central United States.

Exploring the effectiveness of dynamic ensemble selection in the one-versus-one scheme

The One-versus-One (OVO) strategy is one of the most common and effective techniques to deal with multi-class classification problems. The basic idea of an OVO scheme is to divide a multi-class classification problem into several easier-to-solve binary classification problems with considering each possible pair of classes from the original problem, which is then built into a binary classifier by an independent base learner. In this study, we propose a novel methodology which attempts to select a group of base classifiers in each pairwise dataset for each unknown pattern. To implement this, the Dynamic Ensemble Selection (DES) method based on a competence measure is employed to select the most appropriate ensemble in each binary classification problem derived from the OVO decomposition. In order to verify the validity and effectiveness of our proposed method, we carry out a thorough experimental study. We first compare our proposal with several state-of-the-art approaches. Then, we perform the comparison of several well-known aggregation strategies to combine the binary ensemble obtained by Dynamic Ensemble Selection. Finally, we explore whether further improvement can be achieved by considering the competence-based method in OVO scheme. The extracted findings drawn from the empirical analysis are supported by the proper statistical analysis and indicate that there is a positive synergy between the DES method and the Distance-based Relative Competence Weighting (DRCW) approach for the OVO scheme.

Cost-Sensitive back-propagation neural networks with binarization techniques in addressing multi-class problems and non-competent classifiers

Multi-class classification problems can be addressed by using decomposition strategy. One of the most popular decomposition techniques is the One-vs-One (OVO) strategy, which consists of dividing multi-class classification problems into as many as possible pairs of easier-to-solve binary sub-problems. To discuss the presence of classes with different cost, in this paper, we examine the behavior of an ensemble of Cost-Sensitive Back-Propagation Neural Networks (CSBPNN) with OVO binarization techniques for multi-class problems. To implement this, the original multi-class cost-sensitive problem is decomposed into as many sub-problems as possible pairs of classes and each sub-problem is learnt in an independent manner using CSBPNN. Then a combination method is used to aggregate the binary cost-sensitive classifiers. To verify the synergy of the binarization technique and CSBPNN for multi-class cost-sensitive problems, we carry out a thorough experimental study. Specifically, we first develop the study to check the effectiveness of the OVO strategy for multi-class cost-sensitive learning problems. Then, we develop a comparison of several well-known aggregation strategies in our scenario. Finally, we explore whether further improvement can be achieved by using the management of non-competent classifiers. The experimental study is performed with three types of cost matrices and proper statistical analysis is employed to extract the meaningful findings.

Filtering patent maps for visualization of diversification paths of inventors and organizations

In the information science literature, recent studies have used patent databases and patent classification information to construct network maps of patent technology classes. In such a patent technology map, almost all pairs of technology classes are connected, whereas most of the connections between them are extremely weak. This observation suggests the possibility of filtering the patent network map by removing weak links. However, removing links may reduce the explanatory power of the network on inventor or organization diversification. The network links may explain the patent portfolio diversification paths of inventors and inventing organizations. We measure the diversification explanatory power of the patent network map, and present a method to objectively choose an optimal tradeoff between explanatory power and removing weak links. We show that this method can remove a degree of arbitrariness compared with previous filtering methods based on arbitrary thresholds, and also identify previous filtering methods that created filters outside the optimal tradeoff. The filtered map aims to aid in network visualization analyses of the technological diversification of inventors, organizations, and other innovation agents, and potential foresight analysis. Such applications to a prolific inventor (Leonard Forbes) and company (Google) are demonstrated.

Sequence and Phylogenetic Analysis of the Untranslated Promoter Regions for HLA Class I Genes.

Polymorphisms located within the MHC have been linked to many disease outcomes by mechanisms not yet fully understood in most cases. Variants located within untranslated regions of HLA genes are involved in allele-specific expression and may therefore underlie some of these disease associations. We determined sequences extending nearly 2 kb upstream of the transcription start site for 68 alleles from 57 major lineages of classical HLA class I genes. The nucleotide diversity within this promoter segment roughly follows that seen within the coding regions, with HLA-B showing the highest (∼1.9%), followed by HLA-A (∼1.8%), and HLA-C showing the lowest diversity (∼0.9%). Despite its greater diversity, HLA-B mRNA expression levels determined in 178 European Americans do not vary in an allele- or lineage-specific manner, unlike the differential expression levels of HLA-A or HLA-C reported previously. Close proximity of promoter sequences in phylogenetic trees is roughly reflected by similarity of expression pattern for most HLA-A and -C loci. Although promoter sequence divergence might impact promoter activity, we observed no clear link between the phylogenetic structures as represented by pairwise nucleotide differences in the promoter regions with estimated differences in mRNA expression levels for the classical class I loci. Further, no pair of class I loci showed coordinated expression levels, suggesting that distinct mechanisms across loci determine their expression level under nonstimulated conditions. These data serve as a foundation for more in-depth analysis of the functional consequences of promoter region variation within the classical HLA class I loci.

Network analysis reveals crosstalk between autophagy genes and disease genes

Autophagy is a protective and life-sustaining process in which cytoplasmic components are packaged into double-membrane vesicles and targeted to lysosomes for degradation. Accumulating evidence supports that autophagy is associated with several pathological conditions. However, research on the functional cross-links between autophagy and disease genes remains in its early stages. In this study, we constructed a disease-autophagy network (DAN) by integrating known disease genes, known autophagy genes and protein-protein interactions (PPI). Dissecting the topological properties of the DAN suggested that nodes that both autophagy and disease genes (inter-genes), are topologically important in the DAN structure. Next, a core network from the DAN was extracted to analyze the functional links between disease and autophagy genes. The genes in the core network were significantly enriched in multiple disease-related pathways, suggesting that autophagy genes may function in various disease processes. Of 17 disease classes, 11 significantly overlapped with autophagy genes, including cancer diseases, metabolic diseases and hematological diseases, a finding that is supported by the literatures. We also found that autophagy genes have a bridging role in the connections between pairs of disease classes. Altogether, our study provides a better understanding of the molecular mechanisms underlying human diseases and the autophagy process.

An Improved Combination of Spectral and Spatial Features for Vegetation Classification in Hyperspectral Images

Due to the advances in hyperspectral sensor technology, hyperspectral images have gained great attention in precision agriculture. In practical applications, vegetation classification is usually required to be conducted first and then the vegetation of interest is discriminated from the others. This study proposes an integrated scheme (SpeSpaVS_ClassPair_ScatterMatrix) for vegetation classification by simultaneously exploiting image spectral and spatial information to improve vegetation classification accuracy. In the scheme, spectral features are selected by the proposed scatter-matrix-based feature selection method (ClassPair_ScatterMatrix). In this method, the scatter-matrix-based class separability measure is calculated for each pair of classes and then averaged as final selection criterion to alleviate the problem of mutual redundancy among the selected features, based on the conventional scatter-matrix-based class separability measure (AllClass_ScatterMatrix). The feature subset search is performed by the sequential floating forward search method. Considering the high spectral similarity among different green vegetation types, Gabor features are extracted from the top two principal components to provide complementary spatial features for spectral features. The spectral features and Gabor features are stacked into a feature vector and then the ClassPair_ScatterMatrix method is used on the formed vector to overcome the over-dimensionality problem and select discriminative features for vegetation classification. The final features are fed into support vector machine classifier for classification. To verify whether the ClassPair_ScatterMatrix method could well avoid selecting mutually redundant features, the mean square correlation coefficients were calculated for the ClassPair_ScatterMatrix method and AllClass_ScatterMatrix method. The experiments were conducted on a widely used agricultural hyperspectral image. The experimental results showed that (1) the The proposed ClassPair_ScatterMatrix method could better alleviate the problem of selecting mutually redundant features, compared to the AllClass_ScatterMatrix method; (2) compared with the representative mutual information-based feature selection methods, the scatter-matrix-based feature selection methods generally achieved higher classification accuracies, and the ClassPair_ScatterMatrix method especially, produced the highest classification accuracies with respect to both data sets (87.2% and 90.1%); and (3) the proposed integrated scheme produced higher classification accuracy, compared with the decision fusion of spectral and spatial features and the methods only involving spectral features or spatial features. The comparative experiments demonstrate the effectiveness of the proposed scheme.

Two residues determine the product profile of the class II diterpene synthases TPS14 and TPS21 of Tripterygium wilfordii

The medicinal plant Tripterygium wilfordii (Celastraceae) contains a pair of class II diterpene synthases (diTPS) of specialized labdane-type metabolism that, despite remarkably close homology, form strikingly different products. TwTPS21 catalyzes bicyclization of the linear C20 precursor geranylgeranyl diphosphate to ent-copal-8-ol diphosphate, while TwTPS14 forms kolavenyl diphosphate. To determine the amino acid signature controlling the functional divergence of the homologues, we modeled their structures based on an existing crystal structure of the Arabidopsis ent-copalyl diphosphate synthase, archetypal of diTPSs in general metabolism of gibberellin phytohormones. Of the residues differing between TwTPS21 and TwTPS14 two located to the predicted active site, and we hypothesized that these are responsible for the functional differentiation of the enzymes. Using site-directed mutagenesis, we generated a panel of six variants, where one, or both positions were exchanged between the enzymes. In coupled heterologous assays with a corresponding class I diTPS, TwTPS2, complete product interchange was observed in variants with both reciprocal mutations, while substitutions of either residue gave mixed product profiles. Two mutants, TwTPS14:Y265H and TwTPS21:A325V, also produced ent-copalyl diphosphate, highlighting the evolutionary potential of enzymes of this family to drive rapid diversification of plant diterpene biosynthesis through neo-functionalization. Our study contributes to the understanding of structure-function relation in plant class II diTPSs and complements previous mutational studies of Arabidopsis ent-copalyl diphosphate synthase with additional examples from the specialized metabolism of T. wilfordii.

A method for multi-class sentiment classification based on an improved one-vs-one (OVO) strategy and the support vector machine (SVM) algorithm

Multi-class sentiment classification is a valuable research topic with extensive applications; however, studies in the field remain relatively scarce. In the present paper, a method for multi-class sentiment classification based on an improved one-vs-one (OVO) strategy and the support vector machine (SVM) algorithm is proposed. First, an improved OVO strategy is proposed wherein the relative competence weight of each binary classifier is determined according to the K nearest neighbors and the class center of each class in the training sample set concerning the binary classifier. A method for multi-class sentiment classification is proposed based on this improved OVO strategy and the SVM algorithm. After converting the training texts into term feature vectors, the important features (terms) for multi-class sentiment classification are selected using the information gain (IG) algorithm. A binary SVM classifier is then trained on the training feature vectors of each pair of sentiment classes. To identify the sentiment class of a test text, a confidence score matrix of multiple SVM classifiers is constructed based on the results of multiple SVM classifiers. Using this score matrix, the sentiment class of the test text can be determined using the improved OVO strategy. The results of our experimental studies show that the performance of the proposed method is significantly better than that of the existing methods for multi-class sentiment classification.

Dimension Reduction by Minimum Error Minimax Probability Machine

Dimension reduction is frequently adopted as a data preprocessing technique to facilitate data visualization, interpretation, and classification. Traditional dimension reduction methods such as linear discriminant analysis focus on maximizing the overall discrimination between all classes, which may be easily affected by outliers. To overcome this disadvantage, this paper proposes a novel method for multiclass dimension reduction, named dimension reduction by minimum error minimax probability machine (DR-MEMPM). It elaborately ensures that each pair of classes is well separated in the projected subspace by utilizing the separation probability between different pairwise classes. Therefore, it can put more emphasis on those less distinguishable classes, and the learned projection will not be dominated by some “outlier” classes which lie far away from other classes. We evaluate the proposed DR-MEMPM on a number of synthetic and real-world data sets, and show that it outperforms other state-of-the-art dimension reduction methods in terms of visual intuition and classification accuracy, especially when the distances between classes are unevenly distributed.

Classification of Broadcast News Audio Data Employing Binary Decision Architecture

A novel binary decision architecture (BDA) for broadcast news audio classification task is presented in this paper. The idea of developing such architecture came from the fact that the appropriate combination of multiple binary classifiers for two-class discrimination problem can reduce a miss-classification error without rapid increase in computational complexity. The core element of classification architecture is represented by a binary decision (BD) algorithm that performs discrimination between each pair of acoustic classes, utilizing two types of decision functions. The first one is represented by a simple rule-based approach in which the final decision is made according to the value of selected discrimination parameter. The main advantage of this solution is relatively low processing time needed for classification of all acoustic classes. The cost for that is low classification accuracy. The second one employs support vector machine (SVM) classifier. In this case, the overall classification accuracy is conditioned by finding the optimal parameters for decision function resulting in higher computational complexity and better classification performance. The final form of proposed BDA is created by combining four BD discriminators supplemented by decision table. The effectiveness of proposed BDA, utilizing rule-based approach and the SVM classifier, is compared with two most popular strategies for multiclass classification, namely the binary decision trees (BDT) and the One-Against-One SVM (OAOSVM). Experimental results show that the proposed classification architecture can decrease the overall classification error in comparison with the BDT architecture. On the contrary, an optimization technique for selecting the optimal set of training data is needed in order to overcome the OAOSVM.

On Classes of Matrices with Variants of the Diagonal Dominance Property

We study the relations between several classes of matrices with variants of the diagonal dominance property, and identify those classes which form pairs of incomparable classes. For an incomparable pair (X1,X2) of classes of matrices with variants of the diagonal dominance property, we also study the problem of providing sufficient conditions for the matrices in Xi to be in Xj with {i,j}={1,2}. The article is a continuation of a series of articles on the topic and related topics by the author; see [1][2][3][4].

Statistical pair pruning towards target class in learning-based anaphora resolution for Tamil

Anaphora resolution is an important task to be achieved in many natural language understanding (NLU) applications including machine translation. This paper proposes learning-based system to resolve pronouns in Tamil text built around various classification algorithms. To improve learning accuracy, the system is built in two folds. First is feature vector production where mentions are identified, characterised then a feature vectors of lexical, syntactic and semantic features are produced. Next is the pair pruning module where, number of non-target class pairs is reduced by deep statistical analysis of feature vector. Incorporating deeper pair pruning module dramatically increases the f-measure score when compared to training the same models but without the pruning module. On the tourism dataset of TDIL we trained the system with various classification algorithms and obtained encouraging results for a challenging language, Tamil. We discuss how varying the ratio of f-measure, precision and recall is between with and without the pruning module in comparative model.

Fault diagnosis of marine 4-stroke diesel engines using a one-vs-one extreme learning ensemble

This paper proposes a novel approach for intelligent fault diagnosis for stroke Diesel marine engines, which are commonly used in on-road and marine transportation. The safety and reliability of a ship's work rely strongly on the performance of such an engine; therefore, early detection of any type of failure that affects the engine is of crucial importance. Automatic diagnostic systems are of special importance because they can operate continuously in real time, thereby providing efficient monitoring of the engine's performance. We introduce a fully automatic machine learning-based system for engine fault detection. For this purpose, we monitor various signals that are emitted by the engine, and we use them as an input for a pattern classification algorithm. This action is realized by an ensemble of Extreme Learning Machines that work in a decomposition mode. Because we address 14 different faults and a correct operation mode, we must handle a 15-class problem. We tackle this task by binarization in one-vs-one mode, where each Extreme Learning Machine is trained on a pair of classes. Next, Error-Correcting Output Codes are used to reconstruct the original multi-class task. The results from experiments that were conducted on a real-life dataset demonstrate that the proposed approach delivers superior classification accuracy and a low response time in comparison with a number of state-of-the-art methods and thus is a suitable choice for a real-life implementation on board a ship.

NMC: nearest matrix classification – A new combination model for pruning One-vs-One ensembles by transforming the aggregation problem

The One-vs-One strategy is among the most used techniques to deal with multi-class problems in Machine Learning. This way, any binary classifier can be used to address the original problem, since one classifier is learned for each possible pair of classes. As in every ensemble method, classifier combination becomes a vital step in the classification process. Even though many combination models have been developed in the literature, none of them have dealt with the possibility of reducing the number of generated classifiers after the training phase, i.e., ensemble pruning, since every classifier is supposed to be necessary.On this account, our objective in this paper is two-fold: (1) We propose a transformation of the aggregation step, which lead us to a new combination strategy where instances are classified on the basis of the similarities among score-matrices. (2) This fact allows us to introduce the possibility of reducing the number of binary classifiers without affecting the final accuracy. We will show that around 50% of classifiers can be removed (depending on the base learner and the specific problem) and that the confidence degrees obtained by these base classifiers have a strong influence on the improvement in the final accuracy.A thorough experimental study is carried out in order to show the behavior of the proposed approach in comparison with the state-of-the-art combination models in the One-vs-One strategy. Different classifiers from various Machine Learning paradigms are considered as base classifiers and the results obtained are contrasted with the proper statistical analysis.