scholarly journals Towards a Universal Semantic Dictionary

2019 ◽  
Vol 9 (19) ◽  
pp. 4060 ◽  
Author(s):  
Maria Jose Castro-Bleda ◽  
Eszter Iklódi ◽  
Gábor Recski ◽  
Gábor Borbély
Keyword(s):  
Arbitrary Number ◽  
Training Data ◽  
Word Embeddings ◽  
Italian Translation ◽  
Benchmark Data ◽  
First Case ◽  
Comparable Performance ◽  
Baseline System ◽  
Novel Method ◽  
Better Than

A novel method for finding linear mappings among word embeddings for several languages, taking as pivot a shared, multilingual embedding space, is proposed in this paper. Previous approaches learned translation matrices between two specific languages, while this method learns translation matrices between a given language and a shared, multilingual space. The system was first trained on bilingual, and later on multilingual corpora as well. In the first case, two different training data were applied: Dinu’s English–Italian benchmark data, and English–Italian translation pairs extracted from the PanLex database. In the second case, only the PanLex database was used. The system performs on English–Italian languages with the best setting significantly better than the baseline system given by Mikolov, and it provides a comparable performance with more sophisticated systems. Exploiting the richness of the PanLex database, the proposed method makes it possible to learn linear mappings among an arbitrary number of languages.

scholarly journals Towards a Universal Semantic Dictionary

2019 ◽  
Author(s):  
María José Castro-Bleda ◽  
Eszter Iklodi ◽  
Gabor Recski ◽  
Gabor Borbely
Keyword(s):  
Training Data ◽  
Italian Translation ◽  
Benchmark Data ◽  
First Case ◽  
Universal Space ◽  
Comparable Performance ◽  
Baseline System ◽  
Novel Method ◽  
Universal Embedding ◽  
Better Than

A novel method for finding linear mappings among word embeddings for several languages, taking as pivot a shared, universal embedding space, is proposed in this paper. Previous approaches learn translation matrices between two specific languages, but this method learn translation matrices between a given language and a shared, universal space. The system was first trained on bilingual, and later on multilingual corpora as well. In the first case two different training data were applied; Dinu’s English-Italian benchmark data, and English-Italian translation pairs extracted from the PanLex database. In the second case only the PanLex database was used. The system performs on English-Italian languages with the best setting significantly better than the baseline system of Mikolov et al. [1], and it provides a comparable performance with the more sophisticated systems of Faruqui and Dyer [2] and Dinu et al. [3]. Exploiting the richness of the PanLex database, the proposed method makes it possible to learn linear mappings among an arbitrary number of languages.

scholarly journals Triplanar ensemble U-Net model for white matter hyperintensities segmentation on MR images

2020 ◽  
Author(s):  
Vaanathi Sundaresan ◽  
Giovanna Zamboni ◽  
Peter M. Rothwell ◽  
Mark Jenkinson ◽  
Ludovica Griffanti
Keyword(s):  
White Matter ◽  
White Matter Hyperintensities ◽  
Training Data ◽  
Periventricular White Matter ◽  
Mr Images ◽  
Comparable Performance ◽  
Brain Mr Images ◽  
Reliable Quantification ◽  
Anatomical Information ◽  
Better Than

AbstractWhite matter hyperintensities (WMHs) have been associated with various cerebrovascular and neurodegenerative diseases. Reliable quantification of WMHs is essential for understanding their clinical impact in normal and pathological populations. Automated segmentation of WMHs is highly challenging due to heterogeneity in WMH characteristics between deep and periventricular white matter, presence of artefacts and differences in the pathology and demographics of populations. In this work, we propose an ensemble triplanar network that combines the predictions from three different planes of brain MR images to provide an accurate WMH segmentation. Also, the network uses anatomical information regarding WMH spatial distribution in loss functions for improving the efficiency of segmentation and to overcome the contrast variations between deep and periventricular WMHs. We evaluated our method on 5 datasets, of which 3 are part of a publicly available dataset (training data for MICCAI WMH Segmentation Challenge 2017 - MWSC 2017) consisting of subjects from three different cohorts. On evaluating our method separately in deep and periventricular regions, we observed robust and comparable performance in both regions. Our method performed better than most of the existing methods, including FSL BIANCA, and on par with the top ranking deep learning method of MWSC 2017.

scholarly journals Learning with mitigating random consistency from the accuracy measure

2020 ◽  
Vol 109 (12) ◽  
pp. 2247-2281
Author(s):  
Jieting Wang ◽  
Yuhua Qian ◽  
Feijiang Li
Keyword(s):  
Performance Measure ◽  
Data Sets ◽  
Human Beings ◽  
Bayes Risk ◽  
Generalization Bounds ◽  
Benchmark Data ◽  
Accuracy Measure ◽  
Machine Leaning ◽  
Comparable Performance ◽  
Better Than

AbstractHuman beings may make random guesses in decision-making. Occasionally, their guesses may generate consistency with the real situation. This kind of consistency is termed random consistency. In the area of machine leaning, the randomness is unavoidable and ubiquitous in learning algorithms. However, the accuracy (A), which is a fundamental performance measure for machine learning, does not recognize the random consistency. This causes that the classifiers learnt by A contain the random consistency. The random consistency may cause an unreliable evaluation and harm the generalization performance. To solve this problem, the pure accuracy (PA) is defined to eliminate the random consistency from the A. In this paper, we mainly study the necessity, learning consistency and leaning method of the PA. We show that the PA is insensitive to the class distribution of classifier and is more fair to the majority and the minority than A. Subsequently, some novel generalization bounds on the PA and A are given. Furthermore, we show that the PA is Bayes-risk consistent in finite and infinite hypothesis space. We design a plug-in rule that maximizes the PA, and the experiments on twenty benchmark data sets demonstrate that the proposed method performs statistically better than the kernel logistic regression in terms of PA and comparable performance in terms of A. Compared with the other plug-in rules, the proposed method obtains much better performance.

scholarly journals Beyond Cross-Validation—Accuracy Estimation for Incremental and Active Learning Models

2020 ◽  
Vol 2 (3) ◽  
pp. 327-346
Author(s):  
Christian Limberg ◽  
Heiko Wersing ◽  
Helge Ritter
Keyword(s):  
Active Learning ◽  
Cross Validation ◽  
Recognition Task ◽  
Training Data ◽  
Data Sets ◽  
Accuracy Estimation ◽  
Benchmark Data ◽  
Machine Learning Applications ◽  
Novel Method ◽  
Human Teacher

For incremental machine-learning applications it is often important to robustly estimate the system accuracy during training, especially if humans perform the supervised teaching. Cross-validation and interleaved test/train error are here the standard supervised approaches. We propose a novel semi-supervised accuracy estimation approach that clearly outperforms these two methods. We introduce the Configram Estimation (CGEM) approach to predict the accuracy of any classifier that delivers confidences. By calculating classification confidences for unseen samples, it is possible to train an offline regression model, capable of predicting the classifier’s accuracy on novel data in a semi-supervised fashion. We evaluate our method with several diverse classifiers and on analytical and real-world benchmark data sets for both incremental and active learning. The results show that our novel method improves accuracy estimation over standard methods and requires less supervised training data after deployment of the model. We demonstrate the application of our approach to a challenging robot object recognition task, where the human teacher can use our method to judge sufficient training.

scholarly journals Heat Exchanger Network Retrofit of an Oleochemical Plant through a Cost and Energy Efficiency Approach

2021 ◽  
Vol 5 (2) ◽  
pp. 17
Author(s):  
Valli Trisha ◽  
Kai Seng Koh ◽  
Lik Yin Ng ◽  
Vui Soon Chok
Keyword(s):  
Energy Consumption ◽  
Heat Exchanger ◽  
Cost Effective ◽  
Capital Cost ◽  
Global Market ◽  
Annual Saving ◽  
Heat Exchanger Network ◽  
First Case ◽  
Payback Time ◽  
Better Than

Limited research of heat integration has been conducted in the oleochemical field. This paper attempts to evaluate the performance of an existing heat exchanger network (HEN) of an oleochemical plant at 600 tonnes per day (TPD) in Malaysia, in which the emphases are placed on the annual saving and reduction in energy consumption. Using commercial HEN numerical software, ASPEN Energy Analyzer v10.0, it was found that the performance of the current HEN in place is excellent, saving over 80% in annual costs and reducing energy consumption by 1,882,711 gigajoule per year (GJ/year). Further analysis of the performance of the HEN was performed to identify the potential optimisation of untapped heating/cooling process streams. Two cases, which are the most cost-effective and energy efficient, were proposed with positive results. However, the second case performed better than the first case, at a lower payback time (0.83 year) and higher annual savings (0.20 million USD/year) with the addition of one heat exchanger at a capital cost of USD 134,620. The first case had a higher payback time (4.64 years), a lower annual saving (0.05 million USD/year) and three additional heaters at a capital cost of USD 193,480. This research has provided a new insight into the oleochemical industry in which retrofitting the HEN can further reduce energy consumption, which in return will reduce the overall production cost of oleochemical commodities. This is particularly crucial in making the product more competitive in its pricing in the global market.

scholarly journals NLOS Multipath Classification of GNSS Signal Correlation Output Using Machine Learning

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2503
Author(s):  
Taro Suzuki ◽  
Yoshiharu Amano
Keyword(s):  
Machine Learning ◽  
Satellite System ◽  
Training Data ◽  
Support Vector ◽  
Positioning Errors ◽  
Automated Method ◽  
Global Navigation Satellite ◽  
Better Than ◽  
Signal Correlation

This paper proposes a method for detecting non-line-of-sight (NLOS) multipath, which causes large positioning errors in a global navigation satellite system (GNSS). We use GNSS signal correlation output, which is the most primitive GNSS signal processing output, to detect NLOS multipath based on machine learning. The shape of the multi-correlator outputs is distorted due to the NLOS multipath. The features of the shape of the multi-correlator are used to discriminate the NLOS multipath. We implement two supervised learning methods, a support vector machine (SVM) and a neural network (NN), and compare their performance. In addition, we also propose an automated method of collecting training data for LOS and NLOS signals of machine learning. The evaluation of the proposed NLOS detection method in an urban environment confirmed that NN was better than SVM, and 97.7% of NLOS signals were correctly discriminated.

scholarly journals DeepIsoFun: a deep domain adaptation approach to predict isoform functions

2018 ◽  
Vol 35 (15) ◽  
pp. 2535-2544 ◽  
Author(s):  
Dipan Shaw ◽  
Hao Chen ◽  
Tao Jiang
Keyword(s):  
Network Architecture ◽  
Domain Adaptation ◽  
Multiple Instance Learning ◽  
Training Data ◽  
Supplementary Information ◽  
Operating Characteristics ◽  
Neural Network Architecture ◽  
Receiver Operating Characteristics Curve ◽  
Gene Functions ◽  
Better Than

AbstractMotivationIsoforms are mRNAs produced from the same gene locus by alternative splicing and may have different functions. Although gene functions have been studied extensively, little is known about the specific functions of isoforms. Recently, some computational approaches based on multiple instance learning have been proposed to predict isoform functions from annotated gene functions and expression data, but their performance is far from being desirable primarily due to the lack of labeled training data. To improve the performance on this problem, we propose a novel deep learning method, DeepIsoFun, that combines multiple instance learning with domain adaptation. The latter technique helps to transfer the knowledge of gene functions to the prediction of isoform functions and provides additional labeled training data. Our model is trained on a deep neural network architecture so that it can adapt to different expression distributions associated with different gene ontology terms.ResultsWe evaluated the performance of DeepIsoFun on three expression datasets of human and mouse collected from SRA studies at different times. On each dataset, DeepIsoFun performed significantly better than the existing methods. In terms of area under the receiver operating characteristics curve, our method acquired at least 26% improvement and in terms of area under the precision-recall curve, it acquired at least 10% improvement over the state-of-the-art methods. In addition, we also study the divergence of the functions predicted by our method for isoforms from the same gene and the overall correlation between expression similarity and the similarity of predicted functions.Availability and implementationhttps://github.com/dls03/DeepIsoFun/Supplementary informationSupplementary data are available at Bioinformatics online.

An Investigation of the Characteristics of a Bayesian Military Impulse Noise Classifier

2008 ◽  
Author(s):  
Brian Bucci ◽  
Jeffrey Vipperman
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Bayesian Methods ◽  
False Positive ◽  
Impulse Noise ◽  
False Negative ◽  
Training Data ◽  
Maximum Accuracy ◽  
First Case ◽  
Testing Data

In extension of previous methods to identify military impulse noise in the civilian environmental noise monitoring setting by means of a set of computed scalar metrics input to artificial neural network structures, Bayesian methods are investigated to classify the same dataset. Four interesting cases are identified and analyzed: A) Maximum accuracy achieve on training data, B) Maximum overall accuracy on blind testing data, C) Maximum accuracy on testing data with zero false positive detections, D) Maximum accuracy on testing data with zero false negative rejections. The first case is used to illustrative example and the later three represent actual monitoring modes. All of the cases are compared and contrasted to illuminate respective strengths and weaknesses. Overall accuracies of up to 99.8% are observed with no false negative rejections and accuracies of up to 98.4% are also achieved with no false positive detections.

scholarly journals AnnotatorJ: an ImageJ plugin to ease hand-annotation of cellular compartments

2020 ◽  
Author(s):  
Réka Hollandi ◽  
Ákos Diósdi ◽  
Gábor Hollandi ◽  
Nikita Moshkov ◽  
Péter Horváth
Keyword(s):  
Deep Learning ◽  
Training Data ◽  
Automatic Annotation ◽  
Cellular Analysis ◽  
Manual Labour ◽  
Cellular Compartments ◽  
Analysis Quality ◽  
Microscopy Images ◽  
Imagej Plugin ◽  
Better Than

AbstractAnnotatorJ combines single-cell identification with deep learning and manual annotation. Cellular analysis quality depends on accurate and reliable detection and segmentation of cells so that the subsequent steps of analyses e.g. expression measurements may be carried out precisely and without bias. Deep learning has recently become a popular way of segmenting cells, performing unimaginably better than conventional methods. However, such deep learning applications may be trained on a large amount of annotated data to be able to match the highest expectations. High-quality annotations are unfortunately expensive as they require field experts to create them, and often cannot be shared outside the lab due to medical regulations.We propose AnnotatorJ, an ImageJ plugin for the semi-automatic annotation of cells (or generally, objects of interest) on (not only) microscopy images in 2D that helps find the true contour of individual objects by applying U-Net-based pre-segmentation. The manual labour of hand-annotating cells can be significantly accelerated by using our tool. Thus, it enables users to create such datasets that could potentially increase the accuracy of state-of-the-art solutions, deep learning or otherwise, when used as training data.

scholarly journals An approximate method for 1-D simulation of pollution transport in streams with dead zones

2018 ◽  
Vol 66 (4) ◽  
pp. 437-447 ◽  
Author(s):  
Marek Sokáč ◽  
Yvetta Velísková ◽  
Carlo Gualtieri
Keyword(s):  
Approximate Method ◽  
Statistical Evaluation ◽  
Pollution Transport ◽  
Dead Zones ◽  
Advection Diffusion Equation ◽  
Gumbel’S Distribution ◽  
Novel Method ◽  
Substance Transport ◽  
Better Than ◽  
Tracer Experiments

Abstract Analytical solutions describing the 1D substance transport in streams have many limitations and factors, which determine their accuracy. One of the very important factors is the presence of the transient storage (dead zones), that deform the concentration distribution of the transported substance. For better adaptation to such real conditions, a simple 1D approximation method is presented in this paper. The proposed approximate method is based on the asymmetric probability distribution (Gumbel’s distribution) and was verified on three streams in southern Slovakia. Tracer experiments on these streams confirmed the presence of dead zones to various extents, depending mainly on the vegetation extent in each stream. Statistical evaluation confirms that the proposed method approximates the measured concentrations significantly better than methods based upon the Gaussian distribution. The results achieved by this novel method are also comparable with the solution of the 1D advection-diffusion equation (ADE), whereas the proposed method is faster and easier to apply and thus suitable for iterative (inverse) tasks.

Sign in / Sign up

Export Citation Format

Share Document