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.
EEG signal is used to establish a communication channel between brain and control device. A common scenario of machine learning based brain computer interface (BCI) is class wise accuracy is different from overall accuracy. Accuracy for one class is worse than overall accuracy. Because, error rates of all classes are not converging in same rate during training of learner. Active class selection (ACS)...
This paper presents an automated method for seizure detection in EEGs using an increment entropy (IncrEn) and support vector machines (SVMs). The IncrEn is a measure of the complexity of time series, which characterizes both the permutation of values and the temporal order of values. The IncrEn is used to extract features of epileptic EEGs and normal EEGs. The SVMs are employed to classify seizure...
Decision Tree is one of the most popular supervised Machine Learning algorithms; it is also the easiest to understand. But finding an optimal decision tree for a given data is a harder task and the use of multiple performance metrics adds some complexity to the problem of selecting the most appropriate DT.
Active learning has been widely used to select the most informative data for labeling in classification tasks, except for time series classification. The main challenge of active learning in time series classification is to evaluate the informativeness of a time series instance. Specifically, many informativeness metrics have been proposed for traditional active learning, however, none of them is...
Recently, with the growth of registered accounts in Micro-blog, the number of abnormal accounts also increased. The emergence of advertising accounts and other unusual account heavily affects the normal order of Micro-blog, therefore, finding out an efficient method to distinguish and classify advertising accounts plays an important role in the Micro-blog ecosystem. Based on the machine learning method,...
Sequence-to-sequence models with soft attention had significant success in machine translation, speech recognition, and question answering. Though capable and easy to use, they require that the entirety of the input sequence is available at the beginning of inference, an assumption that is not valid for instantaneous translation and speech recognition. To address this problem, we present a new method...
This paper describes an unsupervised method of adapting deep neural networks (DNNs) for sound source localization (SSL). DNNs-based SSL achieves high localization accuracy for sound data that are similar to training data. However, the accuracy deteriorates if a sound source is at an unknown position in unknown reverberant environments. We solve the problem by using unsupervised adaption of the DNNs'...
This paper considers near optimal design of predictive compression system that accounts for packet loss over unreliable networks. Major challenges to address include, propagation of errors due to packet loss through the prediction loop, mismatch between statistics used for design and during operation, and above all a cost function that is fraught with poor local minima. Accurately estimating and minimizing...
Automatically assessing emotional valence in human speech has historically been a difficult task for machine learning algorithms. The subtle changes in the voice of the speaker that are indicative of positive or negative emotional states are often “overshadowed” by voice characteristics relating to emotional intensity or emotional activation. In this work we explore a representation learning approach...
In this work we study variance in the results of neural network training on a wide variety of configurations in automatic speech recognition. Although this variance itself is well known, this is, to the best of our knowledge, the first paper that performs an extensive empirical study on its effects in speech recognition. We view training as sampling from a distribution and show that these distributions...
A multi-stream framework with deep neural network (DNN) classifiers has been applied in this paper to improve automatic speech recognition (ASR) performance in environments with different reverberation characteristics. We propose a room parameter estimation model to determine the stream weights for DNN posterior probability combination with the aim of obtaining reliable log-likelihoods for decoding...
We describe a graph-based semi-supervised learning method for acoustic data that uses a Deep Neural Network (DNN) combined with a stochastic graph-based entropic regularizer to favor smooth solutions over a graph induced by the data. We consider graph embeddings constructed from the input features and also from dimensionality-reduced encodings obtained from the bottleneck layer of a separate deep...
Binary descriptors not only are beneficial for similarity search, they are also capable of serving as discriminant features for classification purpose. In this paper we propose a new algorithm based on cross entropy to learn effective binary descriptors, dubbed CE-Bits, providing an alternative to L-2 and hinge loss learning. Because of the usage of cross entropy, a min-max binary NP-hard problem...
Stacked autoencoders have shown success in generating robust features for images and speech classifications, but there has been limited work in applying stacked autoencoders in signal recognition. In this paper, we study the feasibility of stacked autoencoder based order recognition of continuous phase FSK. The features used for recognition are the approximate entropy (ApEn) of the received signal,...
Big data is a term that defines data set are so large or complex that traditional data processing applications are inadequate. It is also defined by the 3V's i.e. Volume, Velocity and Variety. By volume we mean the enormous amount of data that the organizations collect from variety of sources. Velocity here stands for unprecedented speed via which the data streams in and needs to be handled within...
Short text classification uses a supervised learning process, and it needs a huge amount of labeled data for training. This process consumes a lot of human resources. In traditional supervised learning problems, active learning can reduce the amount of samples that need to be labeled manually. It achieves this goal by selecting the most representative samples to represent the whole training set. Uncertainty...
Developers summarize their changes to code in commit messages.When a message seems “unusual’', however, this puts doubt into the quality of the code contained in the commit. We trained n-gram language models and used cross-entropy as an indicator of commit message “unusualness” of over 120,000 commits from open source projects.Build statuses collected from Travis-CI were used as a proxy for code quality...
In this article, we present our work on classifier to realize a Wireless Capsule Endoscopy (WCE) including a Smart Vision Chip (SVC). Our classifier is based on fuzzy tree and forest of fuzzy trees. We obtain a sensitivity of 92.80% and a specificity of 91.26% with a false detection rate of 8.74% on a large database, that we have constructed, composed of 18910 images containing 3895 polyps from 20...
The paper considers the issue of effective formation of a representative sample to train a neural network of a multilayer perceptron (MLP). As it is known, the key problem of MLP training is the factor space division into the test, validation and training sets. To solve this problem, an approach based on the use of clustering and a Lipschitz constant estimate is put forward. Kohonen's self-organizing...
A novel technique is proposed for categorizing sports events in videos by tracking the positional and angular displacements of the centroid of the moving object in between successive frames. The various sporting events contained in videos are distinguished either by the speed of motion, for instance walking, jogging and running, or by the trajectory made by the human body while in motion, for instance...
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.