The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This paper describes a Support Vector Machine (SVM)-based obstacle recognition system that can recognize both vehicles and pedestrians using bimodal vision. Different techniques were investigated in order to recognize the detected obstacles by the extraction of a compact and pertinent numeric signature from visible and infrared spectrum. A bi-objective optimization (using error classification rate...
In this paper we propose an Iterative Re-Weighted Least Square procedure in order to solve the Support Vector Machines for regression and function estimation. Furthermore, we include a new algorithm to train Support Vector Machines, covering both the proposed approach instead of the quadratic programming part and the most advanced methods to deal with large training data sets. Finally, the performance...
We present a TLD-based vehicle tracking method, which uses HOG that is precomputed in the detection process, and an online SVM re-detector. We perform HOG-based tracking in EHMI. When the tracking fails, the system performs redetection for neighboring regions. Tracked vehicles are reused as positive data for re-learning. Therefore, the proposed system performs robust tracking without additional computation.
Clustering algorithms based on type-I fuzzy set theory have been used for handling overlapping partitioning area over the last few decades. However, these fail to deal with additional degree of fuzziness within the real life datasets, because the membership values of type-I fuzzy set are crisp real numbers. Therefore, since inception, the type-II fuzzy set theory has been studied to address the weakness...
This paper presents an efficient Parkinson disease diagnosis system using Least Squares Twin Support Vector Machine (LSTSVM) and Particle Swarm Optimization (PSO). LSTSVM is a promising binary classifier and has shown better generalization ability and faster computational speed. PSO is used for feature selection and parameter optimization. Parkinson disease dataset is taken from UCI repository. The...
Cross-validation is a commonly used method for evaluating the effectiveness of Support Vector Machines (SVMs). However, existing SVM cross-validation algorithms are not scalable to large datasets because they have to (i) hold the whole dataset in memory and/or (ii) perform a very large number of kernel value computation. In this paper, we propose a scheme to dramatically improve the scalability and...
This paper highlights the importance of measuring systemic risk of commercial banks. Conditional Value-at-Risk (CoVaR) is used to measure the degree of "risk externalities" that a specific bank contributes to the whole banking system. Our analysis not only presents current levels of systemic risk of individual banks but also the changes with time passes. There is some evidence that larger...
The proposed work describes a distribution-based approach for recognizing people in images. The methodology involves pattern classification using first and second order statistics in Principal Component Analysis (PCA) - based clustering framework. Unknown distributions of pedestrian and non-pedestrian patterns are approximated by learning the first and second order statistics of the sample images...
Labeled data, in real world, is quite scarce compared with unlabeled data. Manual annotation is usually expensive and inefficient. Active learning paradigm is used to handle this problem by identifying the most informative instances to annotate. In this paper, we proposed a new active learning algorithm based on nonparallel support vector machine. Numeric experiment shows the effective performance...
Linking multiple data collections to create longitudinal data is an important research problem with multiple applications. Longitudinal data allows analysts to perform studies that would be unfeasible otherwise. In our research we are interested in linking historical census collections to create longitudinal data that would allow tracking people overtime. The goal of the linking is to identify the...
Credit rating prediction using clustering algorithms has become more and more important in the financial literature. Expanding the ideas of [4] and [5], we propose an approach to generate models for automated credit rating prediction based on support vector domain description (SVDD) and linear regression (LR). The models include the prediction for sovereign and corporate bonds. Another advantage is,...
Abstract-Computer vision techniques such as Structurefrom- Motion (SfM) and object recognition tend to fail on scenes with highly reflective objects because the reflections behave differently to the true geometry of the scene. Such image sequences may be treated as two layers superimposed over each other - the nonreflection scene source layer and the reflection layer. However, decomposing the two...
Now a day, the massive amount of data and information (recently termed as “Big Data”) causes accessibility and retrieval problems if poorly managed. This is due to their relational structure which is more complicate, unexplainable, and unanalyzable with simple or traditional methods. The uniform display of these data and information is also difficult due to their diversified formats. Bag of Words...
A fully instrumented surface vessel involving underactuated structure was designed. The problem of straight line course tracking of the underactuated surface vessel at a constant forward speed is addressed. The Serret-Frenet frame is used to define the tracking error, therefore the position tracking errors could be stabilized by stabilizing the single distance error. Least Squares Support Vector Machines...
In order to solve the problem that the output of ball mill pulverizing system is difficult to directly measured in thermal power plant with double inlet and double outlet ball mill pulverizing system which is a large delay, strong nonlinear system. It introduces the pruning method to improve the incremental least square support vector machine's sparsity that based on the incremental least square support...
Assessing reachability for a dynamical system, that is deciding whether a certain state is reachable from a given initial state within a given cost threshold, is a central concept in controls, robotics, and optimization. Direct approaches to assess reachability involve the solution to a two-point boundary value problem (2PBVP) between a pair of states. Alternative, indirect approaches involve the...
This paper proposes average inverter model operating in two complementary modes suitable for microgrid simulation applications. Three phase voltage source inverter (VSI) connects to the microgrid through an LCL low pass filter and operates either in current controlled (CC) or voltage controlled (VC) mode. Models presented here take into account the nonlinear behavior of the switches, delays in the...
In this work a new version of Enhanced Multivariance Products Representation (EMPR) is taken into consideration. Recent researches on the bivariate arrays (i.e., Matrices) have led us to a new scheme which we have called Tridiagonal Matrix Enhanced Multivariate Products Representation (TMEMPR). Therein we have been consecutively using four term EMPR on its bivariate component under different support...
This work is devoted to the decomposition of a univariate function by using very recently developed Tridiagonal Vector Enhanced Multivariance Products Representation (TVEMPR). To this end the target function is expressed as a bilinear form over the power vector of the independent variable and the function's coefficient vector. Both vectors are composed of denumerable infinite number of elements. The...
Satire exposes humanity's vices and foibles through the use of irony, wit, and sometimes sarcasm too. It is also frequently used in online communities. Recognition of satire can help in many NLP applications like dialogue system and review summarization. In this paper we filter online news articles as satirical or true news documents using SVM (Support Vector Machine) classification method combined...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.