Complementarity Of Information Research Articles

Environmental sound classification using temporal-frequency attention based convolutional neural network

Environmental sound classification is one of the important issues in the audio recognition field. Compared with structured sounds such as speech and music, the time–frequency structure of environmental sounds is more complicated. In order to learn time and frequency features from Log-Mel spectrogram more effectively, a temporal-frequency attention based convolutional neural network model (TFCNN) is proposed in this paper. Firstly, an experiment that is used as motivation in proposed method is designed to verify the effect of a specific frequency band in the spectrogram on model classification. Secondly, two new attention mechanisms, temporal attention mechanism and frequency attention mechanism, are proposed. These mechanisms can focus on key frequency bands and semantic related time frames on the spectrogram to reduce the influence of background noise and irrelevant frequency bands. Then, a feature information complementarity is formed by combining these mechanisms to more accurately capture the critical time–frequency features. In such a way, the representation ability of the network model can be greatly improved. Finally, experiments on two public data sets, UrbanSound 8 K and ESC-50, demonstrate the effectiveness of the proposed method.

Feature fusion for inverse synthetic aperture radar image classification via learning shared hidden space

Multi-sensor fusion recognition is a meaningful task in ISAR image recognition. Compared with a single sensor, multi-sensor fusion can provide richer target information, which is conducive to more accurate and robust identification. However, previous deep learning-based fusion methods do not effectively deal with the redundancy and complementarity of information between different sources. In this letter, we construct a shared hidden space to align features from different sources. Accordingly, we design an end-to-end deep fusion framework to fuse dual ISAR images at the feature level. For combining the multi-source information, deep generalised canonical correlation analysis (DGCCA) is used as the loss item to map features extracted from dual input onto the shared hidden space. Moreover, we propose an efficient and lightweight spatial attention module, named united attention module, which can be embedded between dual-stream convolutional neural networks (CNNs) to promote DGCCA optimisation by information interaction. Compared with other deep fusion frameworks, our model obtains the competitive performance in ISAR image fusion for classification.

Aqueous Metabolite Trends for the Progression of Nonalcoholic Fatty Liver Disease in Female Bariatric Surgery Patients by Targeted 1H-NMR Metabolomics.

Determining biomarkers and better characterizing the biochemical progression of nonalcoholic fatty liver disease (NAFLD) remains a clinical challenge. A targeted 1H-NMR study of serum, combined with clinical variables, detected and localized biomarkers to stages of NAFLD in morbidly obese females. Pre-surgery serum samples from 100 middle-aged, morbidly obese female subjects, grouped on gold-standard liver wedge biopsies (non-NAFLD; steatosis; and fibrosis) were collected, extracted, and analyzed in aqueous (D2O) buffer (1H, 600 MHz). Profiled concentrations were subjected to exploratory statistical analysis. Metabolites varying significantly between the non-NAFLD and steatosis groups included the ketone bodies 3-hydroxybutyrate (↓; p = 0.035) and acetone (↓; p = 0.012), and also alanine (↑; p = 0.004) and a putative pyruvate signal (↑; p = 0.003). In contrast, the steatosis and fibrosis groups were characterized by 2-hydroxyisovalerate (↑; p = 0.023), betaine (↓; p = 0.008), hypoxanthine (↓; p = 0.003), taurine (↓; p = 0.001), 2-hydroxybutyrate (↑; p = 0.045), 3-hydroxyisobutyrate (↑; p = 0.046), and increasing medium chain fatty acids. Exploratory classification models with and without clinical variables exhibited overall success rates ca. 75–85%. In the study conditions, inhibition of fatty acid oxidation and disruption of the hepatic urea cycle are supported as early features of NAFLD that continue in fibrosis. In fibrosis, markers support inflammation, hepatocyte damage, and decreased liver function. Complementarity of NMR concentrations and clinical information in classification models is shown. A broader hypothesis that standard-of-care sera can yield metabolomic information is supported.

Multi-Scale Feature Mapping Network for Hyperspectral Image Super-Resolution

Hyperspectral Image (HSI) can continuously cover tens or even hundreds of spectral segments for each spatial pixel. Limited by the cost and commercialization requirements of remote sensing satellites, HSIs often lose a lot of information due to insufficient image spatial resolution. For the high-dimensional nature of HSIs and the correlation between the spectra, the existing Super-Resolution (SR) methods for HSIs have the problems of excessive parameter amount and insufficient information complementarity between the spectra. This paper proposes a Multi-Scale Feature Mapping Network (MSFMNet) based on the cascaded residual learning to adaptively learn the prior information of HSIs. MSFMNet simplifies each part of the network into a few simple yet effective network modules. To learn the spatial-spectral characteristics among different spectral segments, a multi-scale feature generation and fusion Multi-Scale Feature Mapping Block (MSFMB) based on wavelet transform and spatial attention mechanism is designed in MSFMNet to learn the spectral features between different spectral segments. To effectively improve the multiplexing rate of multi-level spectral features, a Multi-Level Feature Fusion Block (MLFFB) is designed to fuse the multi-level spectral features. In the image reconstruction stage, an optimized sub-pixel convolution module is used for the up-sampling of different spectral segments. Through a large number of verifications on the three general hyperspectral datasets, the superiority of this method compared with the existing hyperspectral SR methods is proved. In subjective and objective experiments, its experimental performance is better than its competitors.

Fine-Grained Pests Recognition Based on Truncated Probability Fusion Network via Internet of Things in Forestry and Agricultural Scenes

Accurate identification of insect pests is the key to improve crop yield and ensure quality and safety. However, under the influence of environmental conditions, the same kind of pests show obvious differences in intraclass representation, while the different kinds of pests show slight similarities. The traditional methods have been difficult to deal with fine-grained identification of pests, and their practical deployment is low. In order to solve this problem, this paper uses a variety of equipment terminals in the agricultural Internet of Things to obtain a large number of pest images and proposes a fine-grained identification model of pests based on probability fusion network FPNT. This model designs a fine-grained feature extractor based on an optimized CSPNet backbone network, mining different levels of local feature expression that can distinguish subtle differences. After the integration of the NetVLAD aggregation layer, the gated probability fusion layer gives full play to the advantages of information complementarity and confidence coupling of multi-model fusion. The comparison test shows that the PFNT model has an average recognition accuracy of 93.18% for all kinds of pests, and its performance is better than other deep-learning methods, with the average processing time drop to 61 ms, which can meet the needs of fine-grained image recognition of pests in the Internet of Things in agricultural and forestry practice, and provide technical application reference for intelligent early warning and prevention of pests.

IBNet: Interactive Branch Network for salient object detection

At present, saliency detection methods have achieved gratifying progress benefiting from the development of deep learning. However, the existing methods always fail to make full use of the label information. To address this problem, we focus on the complementarity of salient body information and salient detail information within the labels and propose the Interactive Branch Network (IBNet) in this paper. Generally, IBNet contains three components of Label Redefinition Module (LRM), Information Exchange Module (IEM) and Connected Flow Loss (CFL). These three components all play an enormously important role in the complementary performance of detection. In LRM, enough useful and meaningful heuristic knowledge from the given labels is expanded for dynamic and collaborative supervised learning. In IEM, different derived branches are assigned to collect different types of features for interactive fusion. In CFL, the losses from all connected branches are merged to calculate the total loss. Extensive experiments on benchmark datasets exhibit the effectiveness and efficiency of the proposed method against the state-of-the-art approaches. The source code is publicly available athttps://github.com/xianfangfx/IBNet.

A Novel Robust Low-rank Multi-view Diversity Optimization Model with Adaptive-Weighting Based Manifold Learning

Multi-view clustering has become a hot yet challenging topic, due mainly to the independence of and information complementarity between different views. Although good results are achieved to a certain extent from typical methods including multi-view based k-means clustering, sparse cooperative representation clustering and subspace clustering, they still suffer from several drawbacks or limitations: (1) When each view is sparse decomposed, it still contains some hidden information for mining, such as the structure of samples, the intra-class similarity measure, and the inter-class diversity discrimination, etc. (2) Most of the existing multi-view methods only consider the local features within each view, but fail to effectively balance the importance of and combine information among different views in a diversified way. To tackle these issues, we propose a novel multi-view diversity learning model based on robust bilinear error decomposition (BED). The BED term with a low rank sparse constraint is an improved non-negative matrix factorization (NMF), which is used to extract the hidden structure information in sparse decomposition and useful diversity discrimination information in error matrix. The preservation of local features and selection of important views are achieved by adaptive weighted manifold learning. Furthermore, the Hilbert Schmidt independence criterion is used as a diversity learning term for mutual learning and fusion among views. Finally, the proposed robust low-rank multi-view diversity learning spectral clustering method is evaluated and benchmarked with eight state-of-the-art methods. Experiments in six real datasets have fully validated the significantly improved accuracy and efficiency of the proposed methodology for effective clustering of multi-view images.

Spatial and local environmental factors outweigh geo‐climatic gradients in structuring taxonomically and trait‐based β‐diversity of benthic algae

AbstractAimUnderstanding the variation in biodiversity and its underlying drivers and mechanisms is a core task in biogeography and ecology. We examined (a) the relative contributions of species replacement (i.e., turnover) and richness difference (i.e., nestedness) to taxonomically and trait‐based β‐diversity of stream benthic algae; (b) whether these two facets of β‐diversity are correlated with each other; and (c) the relative contributions of local environmental, geo‐climatic and spatial factors to the two facets of β‐diversity and their components.LocationHun‐Tai River Basin, northeastern China.TaxonStream benthic algae.MethodsA total of 157 sites were sampled. Mantel tests were used to examine the complementarities between the two facets of β‐diversity and their components. Distance‐based redundancy analysis and variation partitioning were utilized to investigate the relative contributions of local environmental, geo‐climatic and spatial factors to each facet of β‐diversity and their components.ResultsWeak correlations between taxonomically and trait‐based β‐diversity and their components were detected, which indicated complementarity of ecological information. Taxonomically based total β‐diversity was largely driven by turnover, whereas trait‐based total β‐diversity was more driven by nestedness. Variation partitioning results indicated that local environmental and spatial factors contributed more than geo‐climatic variables to the total explained variation in taxonomically and trait‐based β‐diversity.Main conclusionsDifferent facets of β‐diversity and their decomposition are important for understanding diversity patterns of benthic algae relative to abiotic factors. A high level of trait‐based convergence among benthic algae communities, despite high taxonomic divergence, demonstrated turnover of species with similar biological traits across our study region. Our study provides a trait‐based insight into stream benthic algae communities, which was less documented by previous freshwater studies that focused on regions undergoing recovery following human disturbances.

Discordant results between conventional newborn screening and genomic sequencing in the BabySeq Project

PurposeNewborn screening (NBS) is performed to identify neonates at risk for actionable, severe, early-onset disorders, many of which are genetic. The BabySeq Project randomized neonates to receive conventional NBS or NBS plus exome sequencing (ES) capable of detecting sequence variants that may also diagnose monogenic disease or indicate genetic disease risk. We therefore evaluated how ES and conventional NBS results differ in this population. MethodsWe compared results of NBS (including hearing screens) and ES for 159 infants in the BabySeq Project. Infants were considered “NBS positive” if any abnormal result was found indicating disease risk and “ES positive” if ES identified a monogenic disease risk or a genetic diagnosis. ResultsMost infants (132/159, 84%) were NBS and ES negative. Only one infant was positive for the same disorder by both modalities. Nine infants were NBS positive/ES negative, though seven of these were subsequently determined to be false positives. Fifteen infants were ES positive/NBS negative, all of which represented risk of genetic conditions that are not included in NBS programs. No genetic explanation was identified for eight infants referred on the hearing screen. ConclusionThese differences highlight the complementarity of information that may be gleaned from NBS and ES in the newborn period.

Continuous Multi-Angle Remote Sensing and Its Application in Urban Land Cover Classification

Because of the limitations of hardware devices, such as the sensors, processing capacity, and high accuracy altitude control equipment, traditional optical remote sensing (RS) imageries capture information regarding the same scene from mostly one single angle or a very small number of angles. Nowadays, with video satellites coming into service, obtaining imageries of the same scene from a more-or-less continuous array of angles has become a reality. In this paper, we analyze the differences between the traditional RS data and continuous multi-angle remote sensing (CMARS) data, and unravel the characteristics of the CMARS data. We study the advantages of using CMARS data for classification and try to capitalize on the complementarity of multi-angle information and, at the same time, to reduce the embedded redundancy. Our arguments are substantiated by real-life experiments on the employment of CMARS data in order to classify urban land covers while using a support vector machine (SVM) classifier. They show the superiority of CMARS data over the traditional data for classification. The overall accuracy may increase up to about 9% with CMARS data. Furthermore, we investigate the advantages and disadvantages of directly using the CMARS data, and how such data can be better utilized through the extraction of key features that characterize the variations of spectral reflectance along the entire angular array. This research lay the foundation for the use of CMARS data in future research and applications.

Information Complementarities and the Dynamics of Transparency Shock Spillovers

We show that information complementarities play an important role in the spillover of transparency shocks. We exploit staggered revelation of financial misconduct by S&P500 firms and find that the implied cost of capital increases for “close” industry peers relative to “distant” peers. Disclosure also increases. The effects are particularly strong when the close peers share common analysts and institutional ownership with the fraudulent firm. While disclosure remains high for the next four years, with sustained disclosure, the cost of equity starts to decrease. Firms’ financing patterns tilt more towards debt financing initially at the expense of equity, but eventually revert.

Endogenous Acquisition of Multi-Dimensional Information in a Strategic Trading Model

We analyze a strategic trading model with endogenous acquisition of multi-dimensional costly private information. We find that integer constraints on the number of informed traders, coupled with the interactions among different types of informed traders, lead to multiplicity and nonexistence of equilibria in the information market. Multiplicity stems either from information substitutability or from information complementarity, and nonexistence stems from the combined effect of information substitutability and information complementarity. Integer constraints also lead to asymmetric equilibria even when the model setup is symmetric. Under integer constraints, the fixed point theorem is no longer applicable and new methodologies are developed to prove the existence of equilibria as well as to find specific solutions for equilibria.

Multidomain Features Fusion for Zero-Shot Learning

Given a novel class instance, the purpose of zero-shot learning (ZSL) is to learn a model to classify the instance by seen samples and semantic information transcending class boundaries. The difficulty lies in how to find a suitable space for zero-shot recognition. The previous approaches use semantic space or visual space as classification space. These methods, which typically learn visual-semantic or semantic-visual mapping and directly exploit the output of the mapping function to measure similarity to classify new categories, do not adequately consider the complementarity and distribution gap of multiple domain information. In this paper, we propose to learn a multidomain information fusion space by a joint learning framework. Specifically, we consider the fusion space as a shared space in which different domain features can be recovered by simple linear transformation. By learning a n-way classifier of fusion space from the seen class samples, we also obtain the discriminative information of the similarity space to make the fusion representation more separable. Extensive experiments on popular benchmark datasets manifest that our approach achieves state-of-the-art performances in both supervised and unsupervised ZSL tasks.

Hyperspectral Image Superresolution Using Spectrum and Feature Context

Deep learning-based hyperspectral image superresolution methods have achieved great success recently. However, most methods utilize 2D or 3D convolution to explore features, and rarely combine the two types of convolution to design networks. Moreover, when the model only contains 3D convolution, almost all the methods take all the bands of hyperspectral image as input to analyze, which requires more memory footprint. To address these issues, we explore a new structure for hyperspectral image superresolution using spectrum and feature context. Inspired by the high similarity among adjacent bands, we design a dual-channel network through 2D and 3D convolution to jointly exploit the information from both single band and adjacent bands, which is different from previous works. Under the connection of depth split, it can effectively share spatial information so as to improve the learning ability of 2D spatial domain. Besides, our method introduces the features extracted from previous band, which contributes to the complementarity of information and simplifies the network structure. Through feature context fusion, it significantly enhances the performance of the algorithm. Extensive evaluations and comparisons on three public datasets demonstrate that our approach produces the state-of-the-art results over the existing approaches.

Adaptive fractional multi-scale edge-preserving decomposition and saliency detection fusion algorithm

Image fusion expands the space–time scope of the detection target by making full use of the information complementarity between images, which is useful for automatic target recognition. An adaptive fractional multi-scale edge-preserving decomposition based on the weighted least square framework is proposed to fuse infrared (IR) and visible (VIS) images. It decomposes images into the base layer and the detail layer. The saliency map is obtained based on adaptive fractional saliency detection of the IR image. This map is beneficial to fuse the base layer and highlights the saliency information significantly. Then, the detail layer is fused by classical choose-max. The experiments indicate that the proposed method outperforms state-of-the-art fusion methods in extracting the target and preserving background information of the IR and VIS images.

Nonlinear Structural Fusion for Multiplex Network

Many real-world complex systems have multiple types of relations between their components, and they are popularly modeled as multiplex networks with each type of relation as one layer. Since the fusion analysis of multiplex networks can provide a comprehensive insight, the structural information fusion of multiplex networks has become a crucial issue. However, most of these existing data fusion methods are inappropriate for researchers to apply to complex network analysis directly. The feature-based fusion methods ignore the sharing and complementarity of interlayer structural information. To tackle this problem, we propose a multiplex network structural fusion (MNSF) model, which can construct a network with comprehensive information. It is composed of two modules: the network feature extraction (NFE) module and the network structural fusion (NSF) module. (1) In NFE, MNSF first extracts a low-dimensional vector representation of a node from each layer. Then, we construct a node similarity network based on embedding matrices and K-D tree algorithm. (2) In NSF, we present a nonlinear enhanced iterative fusion (EIF) strategy. EIF can strengthen high-weight edges presented in one (i.e., complementary information) or more (i.e., shared information) networks and weaken low-weight edges (i.e., redundant information). The retention of low-weight edges shared by all layers depends on the tightness of connections of their K-order proximity. The usage of higher-order proximity in EIF alleviates the dependence on the quality of node embedding. Besides, the fused network can be easily exploited by traditional single-layer network analysis methods. Experiments on real-world networks demonstrate that MNSF outperforms the state-of-the-art methods in tasks link prediction and shared community detection.

Regional Bullying Text Recognition Based on Two-Branch Parallel Neural Networks

Traditional features and pipelined algorithms ignore the subspace semantic information and different information complementarity of regional bullying text when describing and recognizing regional bullying text. In order to solve the above problems, combined with features of Chinese, a regional bullying text recognition algorithm called Two-Branch Parallel Neural Network (TB-PNN) is proposed. First, the word vector, sentence vector, pinyin and tone features extracted by the word embedding technique and the character feature extracted by the Character Graph Convolutional Neural (CGCN). Secondly, TB-PNN is constructed by Multi-Head Self-Attention Mechanism (MHSA), Capsule Network (CapsNet) and Independent Recurrent Neural Network (IndRNN). The left branch was MHSA-CapsNet and the right branch was Multi-MHSA-IndRNN. The algorithm assigns weights to the fused features through MHSA, uses the CapsNet of the left branch to mine the key features with high weight and generates vector tags, and uses the IndRNN of the right branch to capture the subspace semantic information of the key features in the text. The left and right branches form complementary information. Finally, SoftMax classifier is used to realize the accurate recognition of regional bullying text. The experimental results show that TB-PNN algorithm can effectively improve the recognition accuracy of regional bullying text.

Otwarty dostęp w krajach globalnego Południa

Cel/Teza: Artykuł referuje bieżący etap rozwoju otwartego dostępu w krajach globalnego Południa, ze wskazaniem podobieństw i odmienności pomiędzy różnymi obszarami świata, nakreśleniem specyfiki reprezentatywnych krajów lub grupy krajów oraz określeniem przyczyn ograniczenia ich aktywności w tym zakresie.Koncepcja/Metody badań: Termin „globalne Południe”, odnoszący się do mniej rozwiniętych krajów Azji, Pacyfiku, Afryki i Ameryki Łacińskiej, stał się ostatnio popularny w dyskursie akademickim, chociaż nie został prawidłowo zdefiniowany. W artykule różne przypadki jego użycia są analizowane w konfrontacji z terminem „globalna Północ”. Analizie poddane zostały również dane dotyczące otwartego dostępu w krajach lobalnego Południa dostępne w źródłach internetowych.Wyniki i wnioski: Poszczególne obszary globalnego Południa są pod wieloma względami zróżnicowane, choć wszystkie łączy niższy status ekonomiczny i zespół rozmaitych problemów z tego wynikających. Granice pomiędzy Północą i Południem zmieniają się, w pewnych miejscach zaś zacierają. Środowiska naukowe w tych krajach różnią się pod względem potrzeb, dążeń i wyobrażeń na temat tego, jak powinno, ich zdaniem, wyglądać publikowanie prac naukowych. Ze względu na owe specyficzne cechy próba dopasowania narzędzi wypracowanych w naukowych ośrodkach globalnej Północy może okazać się szkodliwa. Kluczowa okazuje się komplementarność informacji (jej złożoność, organizacja i weryfikacja), która, dzięki narzędziom technicznym, staje się dostępna szerszej naukowej społeczności.Oryginalność/Wartość poznawcza: W polskiej literaturze badawczej zagadnienie to pojawia się rzadko. Artykuł zbiera i zestawia dane dotyczące wszystkich obszarów globalnego Południa obecne głównie w literaturze anglojęzycznej.

Deep Multi-Scale Fusion Neural Network for Multi-Class Arrhythmia Detection.

Automated electrocardiogram (ECG) analysis for arrhythmia detection plays a critical role in early prevention and diagnosis of cardiovascular diseases. Extracting powerful features from raw ECG signals for fine-grained diseases classification is still a challenging problem today due to variable abnormal rhythms and noise distribution. For ECG analysis, the previous research works depend mostly on heartbeat or single scale signal segments, which ignores underlying complementary information of different scales. In this paper, we formulate a novel end-to-end Deep Multi-Scale Fusion convolutional neural network (DMSFNet) architecture for multi-class arrhythmia detection. Our proposed approach can effectively capture abnormal patterns of diseases and suppress noise interference by multi-scale feature extraction and cross-scale information complementarity of ECG signals. The proposed method implements feature extraction for signal segments with different sizes by integrating multiple convolution kernels with different receptive fields. Meanwhile, joint optimization strategy with multiple losses of different scales is designed, which not only learns scale-specific features, but also realizes cumulatively multi-scale complementary feature learning during the learning process. In our work, we demonstrate our DMSFNet on two open datasets (CPSC_2018 and PhysioNet/CinC_2017) and deliver the state-of-art performance on them. Among them, CPSC_2018 is a 12-lead ECG dataset and CinC_2017 is a single-lead dataset. For these two datasets, we achieve the F1 score [Formula: see text] and [Formula: see text] which are higher than previous state-of-art approaches respectively. The results demonstrate that our end-to-end DMSFNet has outstanding performance for feature extraction from a broad range of distinct arrhythmias and elegant generalization ability for effectively handling ECG signals with different leads.

A study of continuous space word and sentence representations applied to ASR error detection

This paper presents a study of continuous word representations applied to automatic detection of speech recognition errors. A neural network architecture is proposed, which is well suited to handle continuous word representations, like word embeddings. We explore the use of several types of word representations: simple and combined linguistic embeddings, and acoustic ones associated to prosodic features, extracted from the audio signal. To compensate certain phenomena highlighted by the analysis of the error average span, we propose to model the errors at the sentence level through the use of sentence embeddings. An approach to build continuous sentence representations dedicated to ASR error detection is also proposed and compared to the Doc2vec approach. Experiments are performed on automatic transcriptions generated by the LIUM ASR system applied to the French ETAPE corpus. They show that the combination of linguistic embeddings, acoustic embeddings, prosodic features, and sentence embeddings in addition to more classical features yields very competitive results. Particularly, these results show the complementarity of acoustic embeddings and prosodic information, and show that the proposed sentence embeddings dedicated to ASR error detection achieve better results than generic sentence embeddings.