Complete Feature Set Research Articles

Computing scores of voice quality and speech intelligibility in tracheoesophageal speech for speech stimuli of varying lengths

In this paper, automatic assessment models are developed for two perceptual variables: speech intelligibility and voice quality. The models are developed and tested on a corpus of Dutch tracheoesophageal (TE) speakers. In this corpus, each speaker read a text passage of approximately 300 syllables and two speech therapists provided consensus scores for the two perceptual variables. Model accuracy and stability are investigated as a function of the amount of speech that is made available for speaker assessment (clinical setting). Five sets of automatically generated acoustic-phonetic speaker features are employed as model inputs. In Part I, models taking complete feature sets as inputs are compared to models taking only the features which are expected to have sufficient support in the speech available for assessment. In Part II, the impact of phonetic content and stimulus length on the computer-generated scores is investigated. Our general finding is that a text encompassing circa 100 syllables is long enough to achieve close to asymptotic accuracy.

What Strikes the Strings of Your Heart?–Multi-Label Dimensionality Reduction for Music Emotion Analysis via Brain Imaging

After 20 years of extensive study in psychology, some musical factors have been identified that can evoke certain kinds of emotions. However, the underlying mechanism of the relationship between music and emotion remains unanswered. This paper intends to find the genuine correlates of music emotion by exploring a systematic and quantitative framework. The task is formulated as a dimensionality reduction problem, which seeks the complete and compact feature set with intrinsic correlates for the given objectives. Since a song generally elicits more than one emotion, we explore dimensionality reduction techniques for multi-label classification. One challenging problem is that the hard label cannot represent the extent of the emotion and it is also difficult to ask the subjects to quantize their feelings. This work tries utilizing the electroencephalography (EEG) signal to solve this challenge. A learning scheme called EEG-based emotion smoothing ( ${{\rm E}^2}{\rm S}$ ) and a bilinear multi-emotion similarity preserving embedding (BME-SPE) algorithm are proposed. We validate the effectiveness of the proposed framework on standard dataset CAL-500. Several influential correlates have been identified and the classification via those correlates has achieved good performance. We build a Chinese music dataset according to the identified correlates and find that the music from different cultures may share similar emotions.

Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification.

Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. A test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of HCM patients. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-second, 12-lead ECG signals. Patients are classified as having HCM if the majority of their recorded heartbeats are recognized as characteristic of HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features—both commonly used and newly-developed ones—from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. We also compared the performance of these two classifiers to that obtained by a logistic regression classifier, and the first two methods performed better than logistic regression. The patient-classification precision of random forests and of support vector machine classifiers is close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features.

Evaluate the Interoperability of Document Format: Based on Translation Practice of OOXML and UOF

Taking both OOXML and UOF standards as examples, we empirically evaluate the interoperability of office document formats from the view of translation practice. With the aim of covering the complete feature set of OOXML and UOF, a novel UOF-Open XML Translator is developed in this study. Thorough experiments demonstrate that our translator implements bidirectional conversion of 80.4% features perfectly and 9.9% features with acceptable discrepancy. Regarding the remaining 9.7% features, more efforts would be taken in future work.

A Bayesian non-linear method for feature selection in machine translation quality estimation

We perform a systematic analysis of the effectiveness of features for the problem of predicting the quality of machine translation (MT) at the sentence level. Starting from a comprehensive feature set, we apply a technique based on Gaussian processes, a Bayesian non-linear learning method, to automatically identify features leading to accurate model performance. We consider application to several datasets across different language pairs and text domains, with translations produced by various MT systems and scored for quality according to different evaluation criteria. We show that selecting features with this technique leads to significantly better performance in most datasets, as compared to using the complete feature sets or a state-of-the-art feature selection approach. In addition, we identify a small set of features which seem to perform well across most datasets.

Role of Feature Selection on Leaf Image Classification

The digital images have been studied for image classification, enhancement, image compression and image segmentation purposes. In the present work, it is proposed to study the effects of feature selection algorithm on the predictive classification accuracy of algorithms used for discriminating the different plant leaf images. The process involves extracting the important texture features from the digital images and then subjecting them to feature selection and further classification process. The leaf image features have been extracted by using Gabor texture features and these Gabor features are subjected to Random Forest feature selection algorithm for extracting important texture features. The four classification algorithms like K-Nearest Neighbour, J48, Classification and Regression Trees and Random Forest have been used for classification purpose. This study shows that there is a net improvement in the predictive classification accuracy values, when classification algorithms have been applied on selected features over the complete set of features.

Evaluation of Music Features for PUK Kernel Based Genre Classification

Since music conveys as well as evokes a wealth of emotions in a listener, there has been tremendous research and commercial development to automatically organize music using smart machine learning techniques. In this work, various features are extracted from the music signal for an effective representation to aid in genre classification. The feature set comprises of dynamic, rhythm, tonal, and spectral features comprising a total of 144 features. The size of feature set is further reduced to 39 features using correlation-based feature selection mechanism to remove the correlated features. Support vector machine classifier is used to train the genre classification system with a flexible Pearson Universal Kernel (PUK) that can adapt its behavior to various functions (from linear to Gaussian). The reduced feature set, consisting mostly rhythm and spectral features, significantly outperforms the complete feature set leading to an accuracy of 82% for classifying 5 genres.

Relevance feedback for human motion retrieval using a boosting approach

Content-based human motion retrieval (CBMR) has been more and more important with the rapid growth of motion capture data, but the gap between high-level semantic concepts and low-level features hinders further performance improvement. Relevance feedback is an effective tool to narrow the semantic gap and enhance the retrieval performance. However, as a type of variable-length multivariate time series (VLMTS), motion capture data has its own characteristics including high-dimensionality, demand for elastic matching, and difficulty representing different movements in a uniform feature space, which make it much more challenging to design an effective relevance feedback approach. This paper presents a novel boosting approach for CBMR and the main contributions include three aspects. First, to fit in with the characteristics of VLMTS data and meet the real-time requirement of relevance feedback, the ensemble learning framework RankBoost is introduced and k-nearest neighbors combining with dynamic time warping (KNN-DTW) is employed as its weak ranker. Second, the set of extended Boolean geometry features containing much richer geometry elements and measures is used to represent motion content, and it provides a comparatively complete feature set for designing the weak ranker of RankBoost. Third, to solve the over-fitting problem caused by the small-sample training of relevance feedback, a novel learning objective composed of minimizing empirical ranking loss and minimizing the maximum generalization loss is proposed for RankBoost ensemble learning. Experimental results on CMU database and its extended database verify the effectiveness of the proposed approach.

Identifying Hypertrophic Cardiomyopathy Patients by Classifying Individual Heartbeats from 12-lead ECG Signals.

Test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of patients with hypertrophic cardiomyopathy (HCM) where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-seconds, 12-lead ECG signals. Patients are classified as having HCM if the majority of the heartbeats are recognized as HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features - both commonly used and newly-developed ones - from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. The patient-classification precision and F-measure of both classifiers are close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 304 highly informative features can achieve performance measures comparable to that achieved by using the complete set of features.

Long distance Case assignment and intervention

A common assumption in linguistic theory is that structural Case assignment constitutes a clause-bound, local dependency. Finnish Case assignment is at odds with any such analysis. Here, structural Case assignment penetrates non-finite clause boundaries, adjunct-adverbial boundaries and even noun heads. What constraints such wild behavior has remained a mystery. This article finds that Finnish structural Case assignment is constrained by a peculiar kind of intervention/relativized minimality condition. A distinction between full intervention, where complete feature set is involved, and partial intervention, where only a subset of the relevant features are involved, finds support in our analysis. Starke's (2001) multi-feature intervention analysis will be developed to explain the phenomenon. In addition, the Chomsky–Hiraiwa Multiple-Agree hypothesis finds strong support in this work, but the theory of phases and Phase Impenetrability Condition (PIC) does not.

Feature selection using Principal Component Analysis for massive retweet detection

Social networks become a major actor in massive information propagation. In the context of the Twitter platform, its popularity is due in part to the capability of relaying messages (i.e. tweets) posted by users. This particular mechanism, called retweet, allows users to massively share tweets they consider as potentially interesting for others. In this paper, we propose to study the behavior of tweets that have been massively retweeted in a short period of time. We first analyze specific tweet features through a Principal Component Analysis (PCA) to better understand the behavior of highly forwarded tweets as opposed to those retweeted only a few times. Finally, we propose to automatically detect the massively retweeted messages. The qualitative study is used to select the features allowing the best classification performance. We show that the selection of only the most correlated features, leads to the best classification accuracy (F-measure of 65.7%), with a gain of about 2.4 points in comparison to the use of the complete set of features.

On monitoring and predicting mobile network traffic abnormality

Traffic analysis and traffic abnormality detection are emerged as an efficient way of detecting network attacks in recent years. The existing approaches can be improved by introducing a new model and a new analysis method of network user’s traffic behaviors. The description dimensions to network user’s traffic behaviors in the current approaches are high, resulting in high processing complexity, high delay in differentiating an individual user’s abnormal traffic behavior from massive network data, and low detection rate. To improve the detection rate and efficiency, we develop a new method of establishing user’s traffic behavior analysis system based on a new model of network traffic monitoring. First, we establish a more complete feature set based on the characteristics of network traffic to describe massive network user’s behaviors. Then, we define a feature selection rule based on the relative deviation distance to select the optimized feature set. We use the selected feature set to locate the abnormality moment and the users who produce the abnormal traffic behavior. Finally, a traffic behavior analysis method based on prediction is developed to improve efficiency of the system. This new method is applied to evaluate the mobile users on mobile cloud. The experimental results show that the proposed method has a higher detection rate and lower delay in the analysis of abnormal user’s traffic behavior than that of the existing approaches.

Online feature selection and classification with incomplete data

This paper presents a classification system in which learning, feature selection, and classification for incomplete data are simultaneously carried out in an online manner. Learning is conducted on a predefined model including the class-dependent mean vectors and correlation coefficients, which are obtained by incrementally processing the incoming observations with missing features. A nearest neighbor with a Gaussian mixture model, whose parameters are also estimated from the trained model, is used for classification. When a testing observation is received, the algorithm discards the missing attributes on the observation and ranks the available features by performing feature selection on the model that has been trained so far. The developed algorithm is tested on a benchmark dataset. The effect of missing features for online feature selection and classification are discussed and presented. The algorithm easily converges to the stable state of feature selection with similar accuracy results as those when using the complete and incomplete feature set with up to 50% missing data.

NEAR-OPTIMAL GEOMETRIC FEATURE SELECTION FOR VISUAL SPEECH RECOGNITION

To improve the accuracy of visual speech recognition systems, selection of visual features is of fundamental importance. Prominent features, which are of maximum relevance for speech classification, need to be selected from a large set of extracted visual attributes. Existing methods apply feature reduction and selection techniques on image pixels constituting region-of-interest (ROI) to reduce data dimensionality. We propose application of feature selection methods on geometrical features to select the most dominant physical features. Two techniques, Minimum Redundancy Maximum Relevance (mRMR) and Correlation-based Feature Selection (CFS), have been applied on the extracted visual features. Experimental results show that recognition accuracy is not compromised when a few selected features from the complete visual feature set are used for classification, thereby reducing processing time and storage overheads considerably. Results are compared with performance of principal components obtained by application of Principal Component Analysis (PCA) on our dataset. Our set of selected features outperforms the PCA transformed data. Results show that the center and corner segments of the mouth are major contributors to visual speech recognition. Teeth pixels are shown to be a prominent visual cue. It is also seen that lip width contributes more towards visual speech recognition accuracy as compared to lip height.

Track data hubs enable visualization of user-defined genome-wide annotations on the UCSC Genome Browser

Summary: Track data hubs provide an efficient mechanism for visualizing remotely hosted Internet-accessible collections of genome annotations. Hub datasets can be organized, configured and fully integrated into the University of California Santa Cruz (UCSC) Genome Browser and accessed through the familiar browser interface. For the first time, individuals can use the complete browser feature set to view custom datasets without the overhead of setting up and maintaining a mirror.Availability and implementation: Source code for the BigWig, BigBed and Genome Browser software is freely available for non-commercial use at http://hgdownload.cse.ucsc.edu/admin/jksrc.zip, implemented in C and supported on Linux. Binaries for the BigWig and BigBed creation and parsing utilities may be downloaded at http://hgdownload.cse.ucsc.edu/admin/exe/. Binary Alignment/Map (BAM) and Variant Call Format (VCF)/tabix utilities are available from http://samtools.sourceforge.net/ and http://vcftools.sourceforge.net/. The UCSC Genome Browser is publicly accessible at http://genome.ucsc.edu.Contact: donnak@soe.ucsc.edu

Robust $$k$$ k -mer frequency estimation using gapped $$k$$ k -mers

Oligomers of fixed length, k, commonly known as k-mers, are often used as fundamental elements in the description of DNA sequence features of diverse biological function, or as intermediate elements in the constuction of more complex descriptors of sequence features such as position weight matrices. k-mers are very useful as general sequence features because they constitute a complete and unbiased feature set, and do not require parameterization based on incomplete knowledge of biological mechanisms. However, a fundamental limitation in the use of k-mers as sequence features is that as k is increased, larger spatial correlations in DNA sequence elements can be described, but the frequency of observing any specific k-mer becomes very small, and rapidly approaches a sparse matrix of binary counts. Thus any statistical learning approach using k-mers will be susceptible to noisy estimation of k-mer frequencies once k becomes large. Because all molecular DNA interactions have limited spatial extent, gapped k-mers often carry the relevant biological signal. Here we use gapped k-mer counts to more robustly estimate the ungapped k-mer frequencies, by deriving an equation for the minimum norm estimate of k-mer frequencies given an observed set of gapped k-mer frequencies. We demonstrate that this approach provides a more accurate estimate of the k-mer frequencies in real biological sequences using a sample of CTCF binding sites in the human genome.

Product and Production Process Modeling and Configuration

Product configuration systems are an emerging technology that supports companies in deploying mass customization strategies. Such strategies need to cover the management of the whole customizable product cycle. Adding process modeling and configuration features to a product configurator may improve its ability to assist mass customization development. In this paper, we describe a modeling framework, PRODPROC, that allows one to model both a product and its production process. We first introduce our framework by describing how configurable products are modeled. Then, we describe the main features and capabilities offered to model production processes and to link them with the corresponding products models. The configuration task namely, the procedure that, from a configurable object/activity generates a configured product/process is then analyzed. We also outline a possible CSP-based implementation of a configurator. A comparison with some of the existing systems for product configuration and process modeling emphasizes that none of the considered system/tools offers the complete set of features supported by PRODPROC for interdependent product and process modeling/configuration.

Enhanced SIFT Descriptor Based on Modified Discrete Gaussian-Hermite Moment

The discrete Gaussian-Hermite moment (DGHM) is a global feature representation method that can be applied to square images. We propose a modified DGHM (MDGHM) method and an MDGHM-based scale-invariant feature transform (MDGHM-SIFT) descriptor. In the MDGHM, we devise a movable mask to represent the local features of a non-square image. The complete set of non-square image features are then represented by the summation of all MDGHMs. We also propose to apply an accumulated MDGHM using multi-order derivatives to obtain distinguishable feature information in the third stage of the SIFT. Finally, we calculate an MDGHM-based magnitude and an MDGHM-based orientation using the accumulated MDGHM. We carry out experiments using the proposed method with six kinds of deformations. The results show that the proposed method can be applied to non-square images without any image truncation and that it significantly outperforms the matching accuracy of other SIFT algorithms.

Estimating the Interior Layout of Buildings Using a Shape Grammar to Capture Building Style

An algorithm is described for determining the interior layout of a building given three pieces of information: (1) the footprint of each story; (2) a reasonably complete set of exterior features; and (3) a shape grammar that describes the building style. Essentially, the algorithm prunes a layout tree generated by interpreting the shape grammar with constraints extracted from the footprint and exterior features. The Queen Anne house, commonly located in Pittsburgh, is chosen as the exemplars of the building style. It is shown how a shape-grammar interpreter for the Queen Anne house was developed and applied to the preceding problem. This shape-grammar interpreter provides the basis for developing grammar interpreters for general parametric shapes. For the purposes of illustration and comparison, applications of the approach to two other distinct building styles are briefly described.

An efficient feature selection method for mobile devices with application to activity recognition

This paper presents a feature selection method for data classification, which combines a model-based variable selection technique and a fast two-stage subset selection algorithm. The relationship between a specified (and complete) set of candidate features and the class label is modeled using a non-linear full regression model which is linear-in-the-parameters. The performance of a sub-model measured by the sum of the squared-errors (SSE) is used to score the informativeness of the subset of features involved in the sub-model. The two-stage subset selection algorithm approaches a solution sub-model with the SSE being locally minimized. The features involved in the solution sub-model are selected as inputs to support vector machines (SVMs) for classification. The memory requirement of this algorithm is independent of the number of training patterns. This property makes this method suitable for applications executed in mobile devices where physical RAM memory is very limited. An application was developed for activity recognition, which implements the proposed feature selection algorithm and an SVM training procedure. Experiments are carried out with the application running on a PDA for human activity recognition using accelerometer data. A comparison with an information gain-based feature selection method demonstrates the effectiveness and efficiency of the proposed algorithm.