Complexity Features Research Articles

Consformer: consciousness detection using transformer networks with correntropy-based measures.

Consciousness detection is important in diagnosis and treatment of disorders of consciousness (DOC). Recent studies have demonstrated that electroencephalography (EEG) signals contain effective information for consciousness state evaluation. We propose two novel EEG measures: the spatiotemporal correntropy and the neuromodulation intensity, to reflect the temporal-spatial complexity in brain signals for consciousness detection. Then, we build a pool of EEG measures with different spectral, complexity and connectivity features, and propose Consformer, a transformer network to learn an adaptive optimization of features for different subjects with the attention mechanism. Experiments are carried out using a large dataset of 280 resting-state EEG recordings of DOC patients. Consformer discriminates minimally conscious state (MCS) from vegetative state (VS) with an accuracy of 85.73% and an F1-score of 86.95%, which achieves the state-of-the-art performance.

LogRegX: An Explainable Regression Network for Cross-Well Geophysical Logs Generation

Geophysical logging instruments continuously measure multiple geophysical properties of borehole rocks, thus providing a feasible way to fine borehole geology modelling. Since the missing problem of well logs is inevitable, it is essential to generate the missing logs by the available ones. Recently, a large body of interdisciplinary studies has demonstrated the effectiveness of applying machine learning to solve the missing logs generation problem, under which the training and testing datasets obey the independent and identical distribution (iid) assumption. This assumption, however, is not satisfied in the case of the cross-well missing logs generation task. A standard method to solve the non-iid issue is to map source and target data to a common feature space and then employ Mean Maximum Discrepancy (MMD) to measure domain differences. However, this method suffers from high computational complexity and poor feature explainability when dealing with logs generation tasks. To solve the above problems, we propose an explainable regression network for cross-well geophysical logs generation named LogRegX. LogRegX integrates single-well feature extraction, cross-well feature alignment, and missing logs prediction while maintaining the explainability of logging features. Specifically, LogRegX leverages the gating mechanism to fuse multi-scale logging features to capture the response characteristics of well logs. The learned source and target feature representations are subject to domain discrepancy constraints, measured by Random Fourier Feature transform induced MMD. Additionally, target-domain information retaining mechanism is introduced to maintain the structure of target data so that the transferred features are explainable. Experiments on real-world field data demonstrate the superiority and the explainability of LogRegX over the existing methods.

Classification of Student Graduation using Naïve Bayes by Comparing between Random Oversampling and Feature Selections of Information Gain and Forward Selection

Class-imbalanced data with high attribute dimensions in datasets frequently contribute to issues in a classification process as this can affect algorithms’ performance in the computing process because there are imbalanced numbers of data in each class and irrelevant attributes that must be processed; therefore, this needs for some techniques to overcome the class-imbalanced data and feature selection to reduce data complexity and irrelevant features. Therefore, this study applied random oversampling (ROs) method to overcome the class-imbalanced data and two feature selections (information gain and forward selection) compared to determine which feature selection is superior, more effective and more appropriate to apply. The results of feature selection then were used to classify the student graduation by creating a classification model of Naïve Bayes algorithm. This study indicated an increase in the average accuracy of the Naïve Bayes method without the ROs preprocessing and the feature selection (81.83%), with the ROs (83.84%), with information gain with 3 selected features (86.03%) and forward selection with 2 selected features (86.42%); consequently, these led to increasing accuracy of 4.2% from no pre-processing to information gain and 4.59% from no pre-processing to forward selection. Therefore, the best feature selection was the forward selection with 2 selected features (GPA of the 8th semester and the overall GPA), and the ROs and both feature selections were proven to improve the performance of the Naïve Bayes method.

Syntactic complexity features of science research article introductions: Rhetorical-functional and disciplinary variation perspectives

An increasing body of genre analysis research within English for Academic Purposes (EAP) has attended to the relationship between linguistic features and their rhetorical functions in academic writing. Previous studies along this line have focused primarily on lexical or phraseological features and less on syntactic complexity features, with few exceptions. This study contributes to this body of research by investigating the rhetorical functions of syntactically complex sentences in science research article (RA) introductions, using a corpus containing the introduction sections of 300 published RAs in six science disciplines. The corpus samples were annotated for rhetorical move-steps using an adapted version of the Create a Research Space model (Swales, 2004) and analyzed using five indices tapping into different aspects of syntactic complexity. Results showed that the syntactic complexity indices varied significantly among the rhetorical move-steps but remained stable across hard-pure and hard-applied disciplines. Furthermore, science RA writers employed different types of syntactic complexity features more often for achieving different rhetorical functions. The implications of our findings for academic writing research and pedagogy are discussed.

Configurations of Technology Commercialization: Evidence from Chinese Spin-off Enterprises

Despite numerous studies on understanding effectiveness of technology commercialization, only limited attention is given to the complex relationship inter and inner its conditions, which formed from the policy system between different levels of government networks and the unbalanced regional technology resources in China. Using data from spin-off enterprises, this paper develops an integrated configuration framework for and provide an empirical test of five predictors, namely policy supply, policy coordination, technology source, firm strength, and R&D investment, which capture three contextual features of institutional, organizational, and technological complexity within Chinese technology commercialization context. Confirmatory factor analysis (CFA) and Fuzzy-set qualitative comparative analysis (fs/QCA) of data identified three distinct effective TC paths: The first, dominator, pertained to the effective regional policy supply and powerful enterprise strength, even if the regional technology source is deficient. The second, devotee, characterized by effective policy coordination, intensive R&D investment, and abundant technology source or powerful enterprise. The last, investor, featured by poor policy coordination and poor regional technical resources, while favorable regional policy supply and R&D investment.

Neural complexity is a common denominator of human consciousness across diverse regimes of cortical dynamics

What is the common denominator of consciousness across divergent regimes of cortical dynamics? Does consciousness show itself in decibels or in bits? To address these questions, we introduce a testbed for evaluating electroencephalogram (EEG) biomarkers of consciousness using dissociations between neural oscillations and consciousness caused by rare genetic disorders. Children with Angelman syndrome (AS) exhibit sleep-like neural dynamics during wakefulness. Conversely, children with duplication 15q11.2-13.1 syndrome (Dup15q) exhibit wake-like neural dynamics during non-rapid eye movement (NREM) sleep. To identify highly generalizable biomarkers of consciousness, we trained regularized logistic regression classifiers on EEG data from wakefulness and NREM sleep in children with AS using both entropy measures of neural complexity and spectral (i.e., neural oscillatory) EEG features. For each set of features, we then validated these classifiers using EEG from neurotypical (NT) children and abnormal EEGs from children with Dup15q. Our results show that the classification performance of entropy-based EEG biomarkers of conscious state is not upper-bounded by that of spectral EEG features, which are outperformed by entropy features. Entropy-based biomarkers of consciousness may thus be highly adaptable and should be investigated further in situations where spectral EEG features have shown limited success, such as detecting covert consciousness or anesthesia awareness.

Hierarchical dispersion Lempel–Ziv complexity for fault diagnosis of rolling bearing

The fault information of rolling bearings is generally contained in vibration signals. How to efficiently unearth fault information from the raw signals is the key to detecting and evaluating the health condition of mechanical equipment. Therefore, a hierarchical dispersion Lempel–Ziv complexity (HDLZC) feature extraction method is developed in this paper to improve the accuracy of fault diagnosis. In this method, dispersion theory addresses the deficiency of Lempel–Ziv complexity, and can obtain more fault features from the raw signal. Second, the hierarchical extraction of high- and low-frequency components from time series can improve the ability to describe dynamic features. Simulations and experiments respectively demonstrate the predominance of HDLZC. The experimental results reveal that this method is significantly better than multiscale dispersive Lempel–Ziv complexity, hierarchical Lempel–Ziv complexity, multiscale dispersion entropy, and multiscale permutation entropy in extracting fault information.

Data complexity-based batch sanitization method against poison in distributed learning

The security of Federated Learning (FL)/Distributed Machine Learning (DML) is gravely threatened by data poisoning attacks, which destroy the usability of the model by contaminating training samples, so such attacks are called causative availability indiscriminate attacks. Facing the problem that existing data sanitization methods are hard to apply to real-time applications due to their tedious process and heavy computations, we propose a new supervised batch detection method for poison, which can fleetly sanitize the training dataset before the local model training. We design a training dataset generation method that helps to enhance accuracy and uses data complexity features to train a detection model, which will be used in an efficient batch hierarchical detection process. Our model stockpiles knowledge about poison, which can be expanded by retraining to adapt to new attacks. Being neither attack-specific nor scenario-specific, our method is applicable to FL/DML or other online or offline scenarios.

Feasibility of automated early postnatal sleep staging in extremely and very preterm neonates using dual-channel EEG

ObjectiveTo investigate the feasibility of automated sleep staging based on quantitative analysis of dual-channel electroencephalography (EEG) for extremely and very preterm infants during their first postnatal days. MethodsWe enrolled 17 preterm neonates born between 25 and 30 weeks of gestational age. Three-hour behavioral sleep observations and simultaneous dual-channel EEG monitoring were conducted for each infant within their first 72 hours after birth. Four kinds of representative and complementary quantitative EEG (qEEG) metrics (i.e., bursting, synchrony, spectral power, and complexity) were calculated and compared between active sleep, quiet sleep, and wakefulness. All analyses were performed in offline mode. ResultsIn separate comparison analyses, significant differences between sleep-wake states were found for bursting, spectral power and complexity features. The automated sleep-wake state classifier based on the combination of all qEEG features achieved a macro-averaged area under the curve of receiver operating characteristic of 74.8%. The complexity features contributed the most to sleep-wake state classification. ConclusionsIt is feasible to distinguish between sleep-wake states within the first 72 postnatal hours for extremely and very preterm infants using qEEG metrics. SignificanceOur findings offer the possibility of starting personalized care dependent on preterm infants’ sleep-wake states directly after birth, potentially yielding long-run benefits for their developmental outcomes.

Correlation study between chaos features of physiological signals and emotions based on the information gain method

Clarifying the correlation between the chaos features of physiological signals and emotions plays an important role in improving the accuracy of emotion recognition. In this paper, the correlations of chaos features such as Lyapunov exponent, complexity, box dimension, and approximate entropy of physiological signals with positive, negative, and neutral emotions are analyzed and studied using the information gain method. Based on this, the Bagging-CHAID model is used for emotion recognition validation analysis. The results show that (1) the approximate entropy of physiological signals has the greatest correlation with positive, negative, and neutral emotions, followed by complexity and Lyapunov exponent features, and box dimensional features have the least correlation with the three emotions, thus, the approximate entropy can be chosen to improve the emotion validated based on Bagging-CHAID model, which showed the accuracy of the correlation analysis.

What is life? Active particles tools towards behavioral dynamics in social-biology and economics.

This review paper is devoted to study the conceptual difficulties that mathematics meets when attempting to describe the complexity of living matter focusing on the challenging perspective of developing a mathematical theory for living systems including mutations and selection. The quest starts with the identification of a number of common complexity features of living systems. Then, mathematical structures are derived to include these features, while mathematical models are derived by inserting in the structures models of individual based interactions. Three applications are examined by active particles methods, i.e., models of SARS2-CoV-2 pandemics, models of idiosyncratic learning in open markets and of the dynamics of prices accounting for human behaviors. A critical study, which pervades the whole paper, shows that also economics can be viewed as a behavioral science thus accounting for specific aspects typical of living systems.

TAASSC 프로그램을 활용하여 원어민과 한국인 대학생이 작성한 에세이에 나타난 명사구 복잡성의 특징 비교

This study investigates how syntactic features can be used to distinguish essays written by Korean college students from NS college students. Essays from the ICNALE(International Corpus Network of Asian Learners of English) corpus were analyzed using TAASSC(the Tool for the Automatic Analysis of Syntactic Sophistication and Complexity) program. Two statistical analyses were performed. First, a multivariate analysis of variance(MANOVA) was employed to examine whether there are statistically significant differences in noun phrase complexity features between NS and Korean students. Second, a stepwise discriminant function analysis was performed to classify into two groups(i.e., NS and Korean students) in terms of noun phrase complexity features. Results showed that Korean and NS students significantly differed in terms of various noun phrase complexity indices. The best four predictors for distinguishing Korean students’ essays from NS students’ essays were dependents per object of the preposition, relative clause modifier per nominal, prepositions per object of the preposition, and dependents per nominal. It was shown that NS students produced more syntactically complex essays than Korean students by employing phrasal embedding. Implications of this study for English writing instruction are discussed.

Multi-Defect Detection in Additively Manufactured Lattice Structures Using 3D Electrical Resistance Tomography

Cellular lattice structures possess high strength-to-weight ratios suitable for advanced lightweight engineering applications. However, their quality and mechanical performance can degrade because of defects introduced during manufacturing or in-service. Their complexity and small length scale features make defects difficult to detect using conventional nondestructive evaluation methods. Here we propose a current injection-based method, electrical resistance tomography (ERT), that can be used to detect damaged struts in conductive cellular lattice structures with their intrinsic electromechanical properties. The reconstructed conductivity distributions from ERT can reveal the severity and location of damaged struts without having to probe each strut. However, the low central sensitivity of ERT may result in image artifacts and inaccurate localization of damaged struts. To address this issue, this study introduces an absolute, high throughput, conductivity reconstruction algorithm for 3D ERT. The algorithm incorporates a strut-based normalized sensitivity map to compensate for lower interior sensitivity and suppresses reconstruction artifacts. Numerical simulations and experiments on fabricated representative cellular lattice structures were performed to verify the ability of ERT to quantitatively identify single and multiple damaged struts. The improved performance of this method compared with classical ERT was observed, based on greatly decreased imaging and reconstructed value errors.

Exploring high-order correlations with deep-broad learning for autism spectrum disorder diagnosis.

Recently, a lot of research has been conducted on diagnosing neurological disorders, such as autism spectrum disorder (ASD). Functional magnetic resonance imaging (fMRI) is the commonly used technique to assist in the diagnosis of ASD. In the past years, some conventional methods have been proposed to extract the low-order functional connectivity network features for ASD diagnosis, which ignore the complexity and global features of the brain network. Most deep learning-based methods generally have a large number of parameters that need to be adjusted during the learning process. To overcome the limitations mentioned above, we propose a novel deep-broad learning method for learning the higher-order brain functional connectivity network features to assist in ASD diagnosis. Specifically, we first construct the high-order functional connectivity network that describes global correlations of the brain regions based on hypergraph, and then we use the deep-broad learning method to extract the high-dimensional feature representations for brain networks sequentially. The evaluation of the proposed method is conducted on Autism Brain Imaging Data Exchange (ABIDE) dataset. The results show that our proposed method can achieve 71.8% accuracy on the multi-center dataset and 70.6% average accuracy on 17 single-center datasets, which are the best results compared with the state-of-the-art methods. Experimental results demonstrate that our method can describe the global features of the brain regions and get rich discriminative information for the classification task.

Macular Vascular Complexity Analysis of Diabetes Mellitus by Swept-Source Optical Coherence Tomographic Angiography

Purpose: This study was designed to evaluate the associations between retinal vascular complexity features, including fractal dimension (FD) and blood vessel tortuosity (BVT), and the severity of diabetic retinopathy (DR) by using optical coherence tomographic angiography (OCTA). Methods: In this prospective cross-sectional study, 1,282 ocular-treatment-naive patients with type 2 diabetes mellitus (DM) (1,059 without DR and 223 with DR) registered in the community of Guangzhou, China, were enrolled. OCTA was used to measure FD and BVT in the superficial capillary plexus (SCP) and the deep capillary plexus (DCP). Univariate and multivariate linear regression analyses were performed to analyze the correlation of FD and BVT in different layers with DR severity. Results: In this study, 1,282 patients with DM (1,282 eyes), with a mean age of 64.2 ± 7.8 years, were included. FD in the DCP decreased and BVT in the DCP increased in patients with DR compared with those in patients without DR, even after adjusting for confounding factors (p < 0.05). Trend analysis showed a significant decrease in the FD values as the DR progressed, whereas the BVT progressively increased with worsening DR severity (p < 0.01). The FD in DCP had a statistically significant positive correlation with FD in SCP and a negative correlation with BVT in SCP and BVT in DCP in all of the participants, including the non-DR group, moderate DR group, and severe DR group (p < 0.01). Conclusions: FD and BVT determined using OCTA might be useful parameters for objectively distinguishing DR from non-DR and indicating DR progression.

The interplay of task complexity (± Here-and-Now) and proficiency in shaping written narratives in Spanish as a foreign language

Research on task complexity focuses on the influence of different task features on language learners' production in terms of linguistic complexity, accuracy, fluency and lexis. Within this field, a line of investigation that has attracted much attention is the influence of resource-directing and resource-dispersing features of cognitive task complexity. Notwithstanding, the evidence assembled to date has yielded contradictory results that are not consistent across task type, proficiency level, and task modality. The study attempts to shed some light by examining the role of task complexity along the [± Here-and-Now] dimension in written production, and whether the effects of task complexity are the same at different proficiency levels. To this aim, data were elicited from Swedish high school learners of Spanish as a FL (n = 67) using two different versions of a narrative task. In turn, the participants' proficiency was assessed using a multiple-choice test. The results suggest that task complexity had a statistically significant effect on the four areas of language production. On the other hand, when splitting the sample into proficiency groups, differences across task conditions were found, but only in the higher proficiency group. These results are interpreted in light of Robinson's Cognition Hypothesis and Skehan's Trade-Off Model.

Coupling a neural network technique with CFD simulations for predicting 2-D atmospheric dispersion analyzing wind and composition effects

The Computational Fluid Dynamics (CFD) tool has a remarkable applicability for the prediction of gas dispersion flows by numerically solving the proper governing equations in realistic scenarios. Depending on the problem complexity, undesirably high computational costs can be incurred, which has encouraged the combined use of Machine Learning (ML) seeking to attenuate the CFD simulations requirement for multiple scenario studies. The present work aims at demonstrating the employment the coupling between CFD and the Artificial Neural Network (ANN) algorithm for representative problems in atmospheric dispersion in a preliminary assessment. A limited set of CFD simulation results was used for training neural networks, whose output is given by flow field interpolators, the potential uses of which include digital twin designing and optimization procedures. One possible strategy is the local approach, which treats the network as a transition rule in the scope of Cellular Automata (CA) modeling, allowing it to learn the dynamic behavior of the addressed physics locally. This method gives rise to simpler neural network architectures with closer computing relatively to the CFD calculation. Assessments have been done by predicting, initially, a scalar field time evolution governed by a 1-D advection-diffusion transport equation to verify the method implementation. Subsequently, species concentration distributions were sought in atmospheric dispersion cases from CFD simulations datasets, comprising four case studies followed in the performed analysis, all considering a bidimensional flow domain and a scenario involving methane leaks. The first one indicated an accurate reproduction of subsequent time steps concentration field referring to the displacement of a methane cloud. The second and third cases concerned a plume formation, in transient and steady-state regimes, respectively; their main outcome was the evidence of the CA-ANN methodology's flexibility to address time-dependent and permanent flow simulations interpolation. The last CFD-based case study comprised an additional complexity feature of gas dispersion problems: the wind influence. By redesigning the investigated data-driven approach in terms of ANN's features and labels choice, promising results followed from the analysis with respect to the simultaneous capturing of two global simulation parameters (wind and leakage speeds boundary conditions) in the species concentration field interpolation.

Confronting complexity

Confronting complexity

Complexity measures and features for times series classification

Time series classification is a growing problem in different disciplines due to the progressive digitalization of the world. The best state-of-the-art algorithms focus on performance, seeking the best possible results, leaving interpretability at a second level, if any. Furthermore, interpretable proposals are far from providing competitive results. In this work, focused on time series classification, we propose a new representation of time series based on a robust and complete set of features. This new representation allows extracting more meaningful information on the underlying time series structure to develop effective classifiers whose results are much easier to interpret than current state-of-the-art models. The proposed feature set allows using the traditional vector-based classification algorithms in time series problems, significantly increasing the number of techniques available for this type of problem. To evaluate the performance of our proposal, we have used the state-of-the-art repository of time series classification, UCR, composed of 112 datasets. The experimental results show that through this representation, more interpretable classifiers can be obtained which are competitive. More specifically, they obtain no statistically significant differences from the second and third-best models of the state-of-the-art. Apart from competitive results in accuracy, our proposal is able to improve the model interpretability based on the set of features proposed.

Investigating features in augmented visual feedback for virtual reality rehabilitation of upper-extremity function through isometric muscle control

Previous studies have demonstrated how augmented feedback can accelerate motor learning. Still, how specific feedback features of complexity and intermittency can influence learning a challenging, force-driven motor task remains largely unknown. This study is an initial investigation of how variations in the complexity and intermittency of augmented visual guidance affect the performance of an isometric muscle control task with a computerized platform. This novel platform has been developed to rehabilitate upper-extremity function after neuromuscular dysfunction (e.g., spinal cord injury, stroke) while utilizing: 1) a position-adjustable arm brace for gravity support; 2) a myoelectric command interface; 3) virtual reality (VR) for motor training. Results from this study elucidate new motor control principles and suggest how augmented guidance may be leveraged in designing VR motor rehabilitation programs, which are highly flexible and customizable to individual users. This study demonstrated that simpler and more intermittent feedback typically resulted in better performance (i.e., shorter computerized motion pathlengths). Supplementary results suggested these feedback modes also reduced cognitive loading (i.e., alpha/beta band magnitudes in electroencephalography) but increased physical arousal (i.e., higher skin conductance). In sum, this study indicates that for complex, force-driven tasks, augmented guidance must be presented selectively to accelerate gains in motor performance. This study suggests that simple and intermittent feedback avoids cognitively overwhelming the user while encouraging physical engagement that supports better performance.