Average Accuracy Improvement Research Articles

Class-distinctiveness-based frequency band selection on the Riemannian manifold for oscillatory activity-based BCIs: preliminary results.

Considering user-specific settings is known to enhance Brain-Computer Interface (BCI) performances. In particular, the optimal frequency band for oscillatory activity classification is highly user-dependent and many frequency band selection methods have been developed in the past two decades. However, it is not well studied whether those conventional methods can be efficiently applied to the Riemannian BCIs, a recent family of BCI systems that utilize the non-Euclidean nature of the data unlike conventional BCI pipelines. In this paper, we proposed a novel frequency band selection method working on the Riemannian manifold. The frequency band is selected considering the class distinctiveness as quantified based on the inter-class distance and the intra-class variance on the manifold. An advantage of this method is that the frequency bandwidth can be adjusted for each individual without intensive optimization steps. In a comparative experiment using a public dataset of motor imagery-based BCI, our method showed a substantial improvement in average accuracy over both a fixed broad frequency band and a popular conventional frequency band selection method. In particular, our method substantially improved performances for subjects with initially low accuracies. This preliminary result suggests the importance of developing new user-specific setting algorithms considering the manifold properties, rather than directly applying methods developed prior to the rise of the Riemannian BCIs.

Monitoring of wastewater treatment process based on multi-stage variational autoencoder

The wastewater treatment process (WWTP) is a physical and biochemical reaction process with multi-stage, non-linear and non-Gaussian characteristics. The multi-stage means that the data has diversity in different substages. In the process monitoring of the WWTP, Affinity Propagation (AP) algorithm is first used to divide the data of the WWTP into stages to process the stage characteristics of data. Next, the capability of Variational Auto-Encoder (VAE) monitoring model to restrict the Gaussian distribution in hidden layers is utilized to address the coexistence of non-Gaussian and nonlinearity. Finally, to illustrate the superiority and feasibility, the proposed model is conducted on the Benchmark Simulation Model 1 (BSM1). The experimental result illustrates that the multistage variational autoencoder (M-VAE) can extract the characteristics of data more comprehensively, with an improvement of average monitoring accuracy by 37.8%, 2.7%, 12.7% and 0.52% in 10/8 process faults compared to the state-of-the-art fault monitoring methods Auto-encoder (AE), VAE, Deep Recurrent Neural Network (DRNN), deep recurrent network with high-order statistic information (HSI-DRNN) respectively.

Semantic-Discriminative Mixup for Generalizable Sensor-based Cross-domain Activity Recognition

It is expensive and time-consuming to collect sufficient labeled data to build human activity recognition (HAR) models. Training on existing data often makes the model biased towards the distribution of the training data, thus the model might perform terribly on test data with different distributions. Although existing efforts on transfer learning and domain adaptation try to solve the above problem, they still need access to unlabeled data on the target domain, which may not be possible in real scenarios. Few works pay attention to training a model that can generalize well to unseen target domains for HAR. In this paper, we propose a novel method called Semantic-Discriminative Mixup (SDMix) for generalizable cross-domain HAR. Firstly, we introduce semantic-aware Mixup that considers the activity semantic ranges to overcome the semantic inconsistency brought by domain differences. Secondly, we introduce the large margin loss to enhance the discrimination of Mixup to prevent misclassification brought by noisy virtual labels. Comprehensive generalization experiments on five public datasets demonstrate that our SDMix substantially outperforms the state-of-the-art approaches with 6% average accuracy improvement on cross-person, cross-dataset, and cross-position HAR.

A fuzzy set based approach for effective feature selection

High dimensionality of input dataset is an important issue in machine learning. Though there are a few effective ways of dimensionality reduction, none of them, other than feature selection, preserve the original features of the input dataset. Feature selection assesses relevance of features along with their potential redundancy for dimensionality reduction. This paper proposes a first of its kind fuzzy graph based technique for measuring feature relevance and redundancy. The proposed technique addresses the uncertainty in interpretation of feature relevance and redundancy by defining a fuzzy set based formulation of feature relevance and redundancy. Superior performance of the proposed algorithm is illustrated by simulation experiments with 20 benchmark datasets. The proposed algorithm has demonstrated an average feature reduction of 86% along with a 10% improvement in average classification accuracy compared to the state-of-the-art algorithms.

Semantically accurate super-resolution Generative Adversarial Networks

This work addresses the problems of semantic segmentation and image super-resolution by jointly considering the performance of both in training a Generative Adversarial Network (GAN). We propose a novel architecture and domain-specific feature loss, allowing super-resolution to operate as a pre-processing step to increase the performance of downstream computer vision tasks, specifically semantic segmentation. We demonstrate this approach using Nearmap’s aerial imagery dataset which covers hundreds of urban areas at 5–7 cm per pixel resolution. We show the proposed approach improves perceived image quality as well as quantitative segmentation accuracy across all prediction classes, yielding an average accuracy improvement of 11.8% and 108% at 4× and 32× super-resolution, compared with state-of-the art single-network methods. This work demonstrates that jointly considering image-based and task-specific losses can improve the performance of both, and advances the state-of-the-art in semantic-aware super-resolution of aerial imagery.

Could Prosthesis Use Provide a Competitive Advantage in Darts?

Competitive darts has become increasingly popular over the past few decades, and efforts have been made to have the game recognized as an Olympic sport in the future. The raised profile of the sport and the associated rewards bring up new challenges for the integrity of the game, as athletes are incentivized to exploit rule ambiguities in order to gain competitive advantages. In this research, it was hypothesized that uneven leg lengths and weights, which are comparatively easily realizable in prosthetic limbs, allow players to lean closer to the target and thus improve their throwing accuracy. This hypothesis was tested in a sample of 13 able-bodied subjects who participated in the study, with three sets of throwing trials; one to establish the baseline and two with a longer and heavier trailing leg, respectively. The findings suggest that these modifications are indeed beneficial, resulting in significantly shorter throwing distances and average accuracy improvements of up to 11%. The debate about the potential competitive advantages of prosthesis-wearing Paralympic athletes over their able-bodied peers previously focused on short track running events, where rules have been established that govern the allowable geometry and configuration of sprint prostheses. It appears that comparable regulations should be considered for darts competitions, in order to ensure fair conditions for all participants.

An effective superpixel-based graph convolutional network for small waterbody extraction from remotely sensed imagery

• We develop a water detection method which is particularly efficient in small water. • Graph convolutional network is innovatively used in image semantic segmentation. • Robustness of SG-waterNet is proven over urban, agricultural and mountainous scenes. Small waterbodies sustain susceptible ecosystems and are influenced by variable dynamics associated with human activities and environmental disturbances. Although remote sensing has displayed efficiency in mapping surface waterbodies on a regular basis, the identification of small waterbodies such as ponds or irrigation ditches remains a challenge, as small waterbodies are often confused with other low-reflectivity surfaces. In this study, a superpixel-based graph convolutional network (GCN) for small waterbody extraction (SG-waterNet) is proposed. Specifically, the SG-waterNet method includes a new object-based representation of an image called a superpixel graph. The superpixel graph contains compact spectral and contextual information and can be exploited by the GCN. A deep GCN architecture is used to efficiently preserve small waterbody features and detect surface waterbodies with high completeness and correctness. We tested the proposed approach on a frequently used open-access Gaofen Image Dataset (GID) and Gaofen-1 image from Hubei Province in China (a total of 11,660 km 2 for research). The extraction accuracy of SG-waterNet for small waterbodies (<2 ha) was between 84.31% and 89.77% at the five evaluation sites, and the method extracted waterbodies 300 m 2 and larger with high confidence. Compared with six state-of-the-art methods, SG-waterNet exhibited significant sensitivity to small waterbodies (especially smaller than 100 m 2 ) and detected small waterbody boundaries with the highest completeness and correctness. The average accuracy improvement achieved with SG-waterNet at the evaluation sites ranged from 11.10% to 13.87%. The proposed method is a significant advancement in small waterbody monitoring and can provide promising and practical solutions for real-world applications.

Machine Learning Approach to Community Detection in a High-Entropy Alloy Interaction Network.

There is a growing trend toward the use of interaction network methods and algorithms, including community-based detection methods, in various fields of science. The approach is already used in many applications, for example, in social sciences and health informatics to analyze behavioral patterns during the COVID-19 pandemic, protein–protein networks in biological sciences, agricultural science, economy, and so forth. This paper attempts to build interaction networks based on high-entropy alloy (HEA) descriptors in order to discover HEA communities with similar functionality. In addition, these communities could be leveraged to discover new alloys not yet included in the data set without any experimental laboratory effort. This research has been carried out using two community detection algorithms, the Louvain algorithm and the enhanced particle swarm optimization (PSO) algorithm. The data set, which is used in this paper, includes 90 HEAs and 6 descriptors. The results reveal 13 alloy communities, and the accuracy of the results is validated by the modularity. The experimental results show that the method with the PSO-based community detection algorithm can achieve alloy communities with an average accuracy improvement of 0.26 compared to the Louvain algorithm. Furthermore, some characteristics of HEAs, for example, their phase composition, could be predicted by the extracted communities. Also, the HEA phase composition has been predicted by the proposed method and achieved about 93% precision.

DGNSS Cooperative Positioning in Mobile Smart Devices: A Proof of Concept

Global Navigation Satellite System (GNSS) constitutes the foremost provider for geo-localization in a growing number of consumer-grade applications and services supporting urban mobility. Therefore, low-cost and ultra-low-cost, embedded GNSS receivers have become ubiquitous in mobile devices such as smartphones and consumer electronics to a large extent. However, limited sky visibility and multipath scattering induced in urban areas hinder positioning and navigation capabilities, thus threatening the quality of position estimates. This work leverages the availability of raw GNSS measurements in ultra-low-cost smartphone chipsets and the ubiquitous connectivity provided by modern, low-latency network infrastructures to enable a Cooperative Positioning (CP) framework. A Proof Of Concept is presented that aims at demonstratingdesigned to demonstrate the feasibility of a GNSS-only CP among networked smartphones embedding ultra-low-costequipped with GNSS receivers. The test campaign presented in this study assessed the feasibility of a client-serverthe approach over 4 G/LTE network connectivity. Results demonstrated an overall service availability above 80&#x0025;, and an average accuracy improvement over the 40&#x0025; w.r.t. to the GNSS standalone solution.

CAE‐GAN: A hybrid model for vehicle trajectory prediction

Trajectory prediction of surrounding vehicles is a crucial capability of intelligent driving vehicles. In a scene, a vehicle and its surrounding vehicles constitute an integral system, and the vehicle's future motion trajectory is affected by the actions of surrounding vehicles. The influencing mode and degree are hidden in the relevant historical information of the vehicle and its neighbour vehicle. The existing trajectory prediction methods either do not consider the confidence of the predicted trajectory, or the accuracy requirement is ignored when considering the confidence of the predicted trajectory. In order to address this problem, a mixed Conditional AutoEncoder Generative Adversarial Network (CAE-GAN) model based on the multi-loss function is proposed. The proposed model uses the encoder–decoder structure with a convolutional social pool to extract general features. The generative adversarial networks (GANs) are used to extract the confidence features of the generated trajectories, which enables the proposed model to generate trajectories that are close to the real trajectories. In addition, a classifier structure based on an LSTM network is added to output the probability that the predicted trajectory belongs to a particular lateral maneuver so that the generated trajectory lateral maneuver of the model is consistent with the real trajectory lateral maneuver. The proposed model is evaluated using the publicly available NGSIM US-101 and I-80 datasets, and results show that the accuracy of the proposed model is higher than that of the existing methods. The proposed model achieves an average accuracy improvement of 16.34% in comparison to the most advanced existing models.

Including dominance effects in the prediction model through locus-specific weights on heterozygous genotypes can greatly improve genomic predictive abilities

The dominance effect is considered to be a key factor affecting complex traits. However, previous studies have shown that the improvement of the model, including the dominance effect, is usually less than 1%. This study proposes a novel genomic prediction method called CADM, which combines additive and dominance genetic effects through locus-specific weights on heterozygous genotypes. To the best of our knowledge, this is the first study of weighting dominance effects for genomic prediction. This method was applied to the analysis of chicken (511 birds) and pig (3534 animals) datasets. A 5-fold cross-validation method was used to evaluate the genomic predictive ability. The CADM model was compared with typical models considering additive and dominance genetic effects (ADM) and the model considering only additive genetic effects (AM). Based on the chicken data, using the CADM model, the genomic predictive abilities were improved for all three traits (body weight at 12th week, eviscerating percentage, and breast muscle percentage), and the average improvement in prediction accuracy was 27.1% compared with the AM model, while the ADM model was not better than the AM model. Based on the pig data, the CADM model increased the genomic predictive ability for all the three pig traits (trait names are masked, here designated as T1, T2, and T3), with an average increase of 26.3%, and the ADM model did not improve, or even slightly decreased, compared with the AM model. The results indicate that dominant genetic variation is one of the important sources of phenotypic variation, and the novel prediction model significantly improves the accuracy of genomic prediction.

A federated approach to Android malware classification through Perm-Maps

In the last decades, mobile-based apps have been increasingly used in several application fields for many purposes involving a high number of human activities. Unfortunately, in addition to this, the number of cyber-attacks related to mobile platforms is increasing day-by-day. However, although advances in Artificial Intelligence science have allowed addressing many aspects of the problem, malware classification tasks are still challenging. For this reason, the following paper aims to propose new special features, called permission maps (Perm-Maps), which combine information related to the Android permissions and their corresponding severity levels. Such features have proven to be very effective in classifying different malware families through the usage of a convolutional neural network. Also, the advantages introduced by the Perm-Maps have been enhanced by a training process based on a federated logic. Experimental results show that the proposed approach achieves up to a 3% improvement in average accuracy with respect to J48 trees and Naive Bayes classifier, and up to 16% compared to multi-layer perceptron classifier. Furthermore, the combined use of Perm-Maps and federated logic allows dealing with unbalanced training datasets with low computational efforts.

The Impact of Global Structural Information in Graph Neural Networks Applications

Graph Neural Networks (GNNs) rely on the graph structure to define an aggregation strategy where each node updates its representation by combining information from its neighbours. A known limitation of GNNs is that, as the number of layers increases, information gets smoothed and squashed and node embeddings become indistinguishable, negatively affecting performance. Therefore, practical GNN models employ few layers and only leverage the graph structure in terms of limited, small neighbourhoods around each node. Inevitably, practical GNNs do not capture information depending on the global structure of the graph. While there have been several works studying the limitations and expressivity of GNNs, the question of whether practical applications on graph structured data require global structural knowledge or not remains unanswered. In this work, we empirically address this question by giving access to global information to several GNN models, and observing the impact it has on downstream performance. Our results show that global information can in fact provide significant benefits for common graph-related tasks. We further identify a novel regularization strategy that leads to an average accuracy improvement of more than 5% on all considered tasks.

A multi-feature fusion decoding study for unilateral upper-limb fine motor imagery.

To address the fact that the classical motor imagination paradigm has no noticeable effect on the rehabilitation training of upper limbs in patients after stroke and the corresponding feature extraction algorithm is limited to a single domain, this paper describes the design of a unilateral upper-limb fine motor imagination paradigm and the collection of data from 20 healthy people. It presents a feature extraction algorithm for multi-domain fusion and compares the common spatial pattern (CSP), improved multiscale permutation entropy (IMPE) and multi-domain fusion features of all participants through the use of decision tree, linear discriminant analysis, naive Bayes, a support vector machine, k-nearest neighbor and ensemble classification precision algorithms in the ensemble classifier. For the same subject, the average classification accuracy improvement of the same classifier for multi-domain feature extraction relative to CSP feature results went up by 1.52%. The average classification accuracy improvement of the same classifier went up by 32.87% relative to the IMPE feature classification results. This study's unilateral fine motor imagery paradigm and multi-domain feature fusion algorithm provide new ideas for upper limb rehabilitation after stroke.

Efficient Stack Distance Approximation Based on Workload Characteristics

A stack distance of a reference is the depth from which the reference must be extracted from a stack. It has been widely applied to a variety of applications utilizing temporal locality information. However, calculating an exact stack distance requires significant computational complexity that is impractical for online production systems. This paper proposes an efficient algorithm that approximates a stack distance based on workload characteristics. The proposed algorithm utilizes two simple reference metrics: inter-reference distance and a quasi-unique reference count. More importantly, its approximation algorithm requires surprisingly simple calculation. Consequently, it significantly reduces computational overhead while exhibiting exceptional accuracy and runs extremely fast. Moreover, it allows configurable parameters for performance optimization by making optimal use of workload characteristics. Our extensive experiments with diverse realistic workloads demonstrate our stack distance approximation algorithm outperforms the current state-of-the-art algorithm with higher accuracy (3.7&#x00D7;), lower memory footprint (3.05&#x00D7;) and fast execution (less than 0.05 seconds) on average. In addition, with configuration parameters, our refined algorithm can achieve a 28% average additional accuracy improvement.

Celestial Spectrum Velocimetry With Non-Linear Fourier Phase Shift and Its CRLB

To solve the problem of the non-linear Fourier phase shift caused by the wavelength shift in the celestial spectrum velocimetry, a celestial spectrum velocimetry method based on non-uniform discrete Fourier transform and compressed sensing is proposed. First, according to the properties of the non-uniform discrete Fourier transform with non-periodic celestial spectra, the non-linear relationship between the spectral wavelength shift and the Fourier phase shift is deduced theoretically. And then, the discrete cosine transform is used to decompose and reconstruct the spectrum to construct the phase dictionary of the non-periodic spectrum. In this dictionary, the spectral wavelength shift has a non-linear relationship with the Fourier phase shift. Finally, the non-uniform Fourier transform matrix is used as the measurement matrix. And the Doppler wavelength shift is estimated by using a non-linear and super-resolution matching algorithm to estimate the velocimetry. Experimental results show that when the spacecraft’s velocity is high, the celestial spectrum velocimetry method with non-linear Fourier phase shift (CSV-NFPS) can effectively measure the velocity from the spectrum, and its accuracy approaches the theoretical Cramer-Rao lower bounds (CRLB). Compared with the traditional Taylor method, the CSV-NFPS has an average accuracy improvement of 16.18% and a computational complexity reduction of 25.07%. In short, our method eliminate the non-linear bias in the celestial spectrum velocimetry and reduce a certain computational complexity.

A Two-Stage Adaptation Network (TSAN) for Remote Sensing Scene Classification in Single-Source-Mixed-Multiple-Target Domain Adaptation (S²M²T DA) Scenarios

Over the past decade, domain adaptation (DA) algorithms have been proposed to address domain gap problems as they do not need any interpretation in the target domain. However, most existing efforts focus on scenarios with only one source domain and one target domain. In this article, we explore the scenario with one source domain and mixed multiple target domains for remote sensing applications and propose a new algorithm, named the two-stage adaptation network (TSAN). First, we utilize the adversarial learning approach to confuse the classifier to discriminate between the source domain and the whole mixed-multiple-target domain. Second, we adopt self-supervised learning to divide the mixed-multiple-target domain with automated generation of “pseudo”-domain labels, which guides our network to learn intrinsic features of multiple target domains. Finally, these two steps are combined as an iterative procedure. We integrate a test dataset that includes five remote sensing datasets and ten classes. Our method achieves an average accuracy of 63.25% and 73.68% with two typical backbones, considerably outperforming other DA methods with an average accuracy improvement of 4.84%–20.19% and 9.06%–17.04%, respectively. Furthermore, we identify the negative transfer effect in existing mainstream DA methods in remote sensing image classification with multiple different domains.

Determination of biomarkers from microarray data using graph neural network and spectral clustering

In bioinformatics, the rapid development of gene sequencing technology has produced an increasing amount of microarray data. This type of data shares the typical characteristics of small sample size and high feature dimensions. Searching for biomarkers from microarray data, which expression features of various diseases, is essential for the disease classification. feature selection has therefore became fundemental for the analysis of microarray data, which designs to remove irrelevant and redundant features. There are a large number of redundant features and irrelevant features in microarray data, which severely degrade the classification effectiveness. We propose an innovative feature selection method with the goal of obtaining feature dependencies from a priori knowledge and removing redundant features using spectral clustering. In this paper, the graph structure is firstly constructed by using the gene interaction network as a priori knowledge, and then a link prediction method based on graph neural network is proposed to enhance the graph structure data. Finally, a feature selection method based on spectral clustering is proposed to determine biomarkers. The classification accuracy on DLBCL and Prostate can be improved by 10.90% and 16.22% compared to traditional methods. Link prediction provides an average classification accuracy improvement of 1.96% and 1.31%, and is up to 16.98% higher than the published method. The results show that the proposed method can have full use of a priori knowledge to effectively select disease prediction biomarkers with high classification accuracy.

Inferring Users' Social Roles with a Multi-Level Graph Neural Network Model.

Users of social networks have a variety of social statuses and roles. For example, the users of Weibo include celebrities, government officials, and social organizations. At the same time, these users may be senior managers, middle managers, or workers in companies. Previous studies on this topic have mainly focused on using the categorical, textual and topological data of a social network to predict users’ social statuses and roles. However, this cannot fully reflect the overall characteristics of users’ social statuses and roles in a social network. In this paper, we consider what social network structures reflect users’ social statuses and roles since social networks are designed to connect people. Taking an Enron email dataset as an example, we analyzed a preprocessing mechanism used for social network datasets that can extract users’ dynamic behavior features. We further designed a novel social network representation learning algorithm in order to infer users’ social statuses and roles in social networks through the use of an attention and gate mechanism on users’ neighbors. The extensive experimental results gained from four publicly available datasets indicate that our solution achieves an average accuracy improvement of 2% compared with GraphSAGE-Mean, which is the best applicable inductive representation learning method.

Unsupervised domain adaptation model for lesion detection in retinal OCT images

Background and objective. Optical coherence tomography (OCT) is one of the most used retinal imaging modalities in the clinic as it can provide high-resolution anatomical images. The huge number of OCT images has significantly advanced the development of deep learning methods for automatic lesion detection to ease the doctor’s workload. However, it has been frequently revealed that the deep neural network model has difficulty handling the domain discrepancies, which widely exist in medical images captured from different devices. Many works have been proposed to solve the domain shift issue in deep learning tasks such as disease classification and lesion segmentation, but few works focused on lesion detection, especially for OCT images. Methods. In this work, we proposed a faster-RCNN based, unsupervised domain adaptation model to address the lesion detection task in cross-device retinal OCT images. The domain shift is minimized by reducing the image-level shift and instance-level shift at the same time. We combined a domain classifier with a Wasserstein distance critic to align the shifts at each level. Results. The model was tested on two sets of OCT image data captured from different devices, obtained an average accuracy improvement of more than 8% over the method without domain adaptation, and outperformed other comparable domain adaptation methods. Conclusion. The results demonstrate the proposed model is more effective in reducing the domain shift than advanced methods.