Unidirectional Network Research Articles

Controlling network ensembles

The field of optimal control typically requires the assumption of perfect knowledge of the system one desires to control, which is an unrealistic assumption for biological systems, or networks, typically affected by high levels of uncertainty. Here, we investigate the minimum energy control of network ensembles, which may take one of a number of possible realizations. We ensure the controller derived can perform the desired control with a tunable amount of accuracy and we study how the control energy and the overall control cost scale with the number of possible realizations. Our focus is in characterizing the solution of the optimal control problem in the limit in which the systems are drawn from a continuous distribution, and in particular, how to properly pose the weighting terms in the objective function. We verify the theory in three examples of interest: a unidirectional chain network with uncertain edge weights and self-loop weights, a network where each edge weight is drawn from a given distribution, and the Jacobian of the dynamics corresponding to the cell signaling network of autophagy in the presence of uncertain parameters.

A Graph Convolutional Stacked Bidirectional Unidirectional-LSTM Neural Network for Metro Ridership Prediction

Timely precise metro ridership forecasting is helpful to reveal real-time traffic demand, which is a crucial but challenging task in modern traffic management. Given the complex spatial correlation and temporal variation of riding behaviour in a metro system, deep learning algorithms have been widely applied owing to their superior performance in capturing spatio-temporal features. However, current deep learning models utilize regular convolutional operations, which can barely provide satisfactory accuracy due to either the ignorance of realistic topology of a traffic network or insufficiency in capturing representative spatiotemporal patterns. To further improve the accuracy in metro ridership prediction, this study proposes a parallel-structured deep learning model that consists of a Graph Convolution Network and a stacked Bidirectional unidirectional Long short-term Memory network (GCN-SBULSTM). The GCN module regards a metro network as a structured graph, and a K-hop matrix, which integrates the travel distance, population flow, and adjacency, is introduced to capture the dynamic spatial correlation among metro stations. The SBULSTM module considers both backward and forward states of ridership time series and can learn complex temporal features with stacked recurrent layers. Experiments are conducted on three real-life metro ridership datasets to demonstrate the effectiveness of the proposed model. Compared with state-of-the-art prediction models, GCN-SBULSTM presents better performance in multiple scenarios and largely enhances the efficiencies of training processes.

Context-driven Bengali Text Generation using Conditional Language Model

Text generation is a rapidly evolving field of Natural Language Processing (NLP) with larger Language models proposed very often setting new state-of-the-art. These models are exorbitantly effective in learning the representation of words and their internal coherence in a particular language. However, an established context-driven, end to end text generation model is very rare, even more so for the Bengali language. In this paper, we have proposed a Bidirectional gated recurrent unit (GRU) based architecture that simulates the conditional language model or the decoder portion of the sequence to sequence (seq2seq) model and is further conditioned upon the target context vectors. We have explored several ways of combining multiple context words into a fixed dimensional vector representation that is extracted from the same GloVe language model which is used to generate the embedding matrix. We have used beam search optimization to generate the sentence with the maximum cumulative log probability score. In addition, we have proposed a human scoring based evaluation metric and used it to compare the performance of the model with unidirectional LSTM and GRU networks. Empirical results prove that the proposed model performs exceedingly well in producing meaningful outcomes depicting the target context. The experiment leads to an architecture that can be applied to an extensive domain of context-driven text generation based applications and which is also a key contribution to the NLP based literature of the Bengali language.

Classification of Arabic healthcare questions based on word embeddings learned from massive consultations: a deep learning approach

Automated question classification is a fundamental component of automated question-answering systems, which plays a critical role in promoting medical and healthcare services. Developing an automated question classification system depends heavily on natural language processing and data mining techniques. Question classification methods based on classical machine learning techniques face limitations in capturing the hidden relationships of features, as well as, handling complex languages and very large-scale datasets. Therefore, this paper proposes a deep learning approach for question classification, since deep learning methods have the powerful capability to extract implicit, hidden relationships and automatically generate dense representations of features. The proposed question classification model depends on unidirectional and bidirectional long short-term memory networks (LSTM and BiLSTM), which essentially developed to handle the Arabic language in the field of healthcare. The features are represented and created using a domain-specific word embedding model (Word2Vec) that is constructed by training around 1.5 million medical consultations from Altibbi company. Altibbi is a telemedicine company that is used as a case study and a source for curating and collecting the data. The proposed deep learning approach is a multi-class classification algorithm that automatically labels and maps the questions into 15 categories of medical specialities. The proposed deep learning model is evaluated using several evaluation metrics, including accuracy, precision, recall, and F1-score. Markedly, the proposed model achieved a superb classification capacity in terms of classification accuracy rate, which gained 87.2%.

Lag synchronization of coupled time-delayed FitzHugh\u2013Nagumo neural networks via feedback control

Synchronization plays a significant role in information transfer and decision-making by neurons and brain neural networks. The development of control strategies for synchronizing a network of chaotic neurons with time delays, different direction-dependent coupling (unidirectional and bidirectional), and noise, particularly under external disturbances, is an essential and very challenging task. Researchers have extensively studied the synchronization mechanism of two coupled time-delayed neurons with bidirectional coupling and without incorporating the effect of noise, but not for time-delayed neural networks. To overcome these limitations, this study investigates the synchronization problem in a network of coupled FitzHugh–Nagumo (FHN) neurons by incorporating time delays, different direction-dependent coupling (unidirectional and bidirectional), noise, and ionic and external disturbances in the mathematical models. More specifically, this study investigates the synchronization of time-delayed unidirectional and bidirectional ring-structured FHN neuronal systems with and without external noise. Different gap junctions and delay parameters are used to incorporate time-delay dynamics in both neuronal networks. We also investigate the influence of the time delays between connected neurons on synchronization conditions. Further, to ensure the synchronization of the time-delayed FHN neuronal networks, different adaptive control laws are proposed for both unidirectional and bidirectional neuronal networks. In addition, necessary and sufficient conditions to achieve synchronization are provided by employing the Lyapunov stability theory. The results of numerical simulations conducted for different-sized multiple networks of time-delayed FHN neurons verify the effectiveness of the proposed adaptive control schemes.

Oriented diameter of star graphs

An orientation of an undirected graph G is an assignment of exactly one direction to each edge of G . Converting two-way traffic networks to one-way traffic networks and bidirectional communication networks to unidirectional communication networks are practical instances of graph orientations. In these contexts minimising the diameter of the resulting oriented graph is of prime interest. The n -star network topology was proposed as an alternative to the hypercube network topology for multiprocessor systems by Akers and Krishnamurthy (1989). The n -star graph S n consists of n ! vertices, each labelled with a distinct permutation of [ n ] . Two vertices are adjacent if their labels differ exactly in the first and one other position. S n is an ( n − 1 ) -regular, vertex-transitive graph with diameter ⌊ 3 ( n − 1 ) / 2 ⌋ . Orientations of S n , called unidirectional star graphs and distributed routing protocols over them were studied by Day and Tripathi (1993) and Fujita (2013). Fujita showed that the (directed) diameter of this unidirectional star graph S n ⃗ is at most ⌈ 5 n / 2 ⌉ + 2 . In this paper, we propose a new distributed routing algorithm for the same S n ⃗ analysed by Fujita, which routes a packet from any node s to any node t at an undirected distance d from s using at most min { 4 d + 4 , 2 n + 4 } hops. This shows that the (directed) diameter of S n ⃗ is at most 2 n + 4 . We also show that the diameter of S n ⃗ is at least 2 n when n ≥ 7 , thereby showing that our upper bound is tight up to an additive factor.

An Empirical Analysis of Corporate Financial Management Risk Prediction Based on Associative Memory Neural Network.

As a human brain-like computational model that can reflect the cognitive function of the brain, the problem of dynamic analysis of associative memory neural networks has attracted the attention of scholars. This paper combines associative memory neural networks with enterprise financial management risks, studies the synchronization control and stability analysis problems of unidirectional associative memory-like human brain amnestic neural networks with perturbation and mixed time-varying time lags, proposes a bidirectional associative memory-like brain stochastic amnestic neural network model with mixed time-varying time lags, designs a discrete-time sampling control strategy based on the model, and studies various types of recent financial risks. Based on the early warning research, based on the associative memory neural network method, we propose to reconstruct the risk categories, including improving the enterprise risk management system, enhancing the awareness of financial risk management from top to bottom, and strengthening the core competitiveness of the enterprise itself and control measures for financing, investment, operation, and cash flow risks.

Modeling techniques for decentralized energy systems applied in smart grids

Historically, power grids have emerged as the most economical way to match diversified generation resources. After the Second World War, all advanced countries chose to develop a centralized electrical system to transport energy throughout the national territory and then distribute it to the various points of consumption. In this centralized management, production is adapted to an ever-increasing demand, driven by economic and demographic developments. Gradually, the consumption profile has changed: the development of electrical uses (particularly heating and air conditioning), leading to consumption peaks that are increasingly difficult to satisfy. The appearance of the electric vehicle reinforces these growing imbalances between supply and demand. Thus, the production profile has gradually changed: production has moved closer to the places of consumption but has become more variable, such as wind and solar energy (known as intermittent sources of energy production). In this transforming energy view, these historical centralized and unidirectional networks reach saturation and need to be modernized by turning to a decentralized model. In this perspective, the development of smart grids (SG) is taking place. This paper aims to define decentralized energy systems that provide various benefits and cause significant challenges. Finally, we propose several techniques that are highly efficient in modeling and controlling smart grid systems in order to help decision-makers to address complex problems.

Shortest Path Algorithm in Dynamic Restricted Area Based on Unidirectional Road Network Model.

Accurate and fast path calculation is essential for applications such as vehicle navigation systems and transportation network routing. Although many shortest path algorithms for restricted search areas have been developed in the past ten years to speed up the efficiency of path query, the performance including the practicability still needs to be improved. To settle this problem, this paper proposes a new method of calculating statistical parameters based on a unidirectional road network model that is more in line with the real world and a path planning algorithm for dynamically restricted search areas that constructs virtual boundaries at a lower confidence level. We conducted a detailed experiment on the proposed algorithm with the real road network in Zhengzhou. As the experiment shows, compared with the existing algorithms, the proposed algorithm improves the search performance significantly in the condition of optimal path under the premise of ensuring the optimal path solution.

Persian speech recognition using deep learning

Up to now, various methods are used for Automatic Speech Recognition (ASR), and among which the Hidden Markov Model (HMM) and Artificial Neural Networks (ANNs) are the most important ones. One of the existing challenges is increasing the accuracy and efficiency of these systems. One way to enhance the accuracy of them is by improving the acoustic model (AM). In this paper, for the first time, the combination of deep belief network (DBN), for extracting features of speech signals, and Deep Bidirectional Long Short-Term Memory (DBLSTM) with Connectionist Temporal Classification (CTC) output layer is used to create an AM on the Farsdat Persian speech data set. The obtained results show that the use of a deep neural network (DNN) compared to a shallow network improves the results. Also, using the bidirectional network increases the accuracy of the model in comparison with the unidirectional network, in both deep and shallow networks. Comparing obtained results with the HMM and Kaldi-DNN indicates that using DBLSTM with features extracted from the DBN increases the accuracy of Persian phoneme recognition.

Virtual UV Fluorescence Microscopy from Hematoxylin and Eosin Staining of Liver Images Using Deep Learning Convolutional Neural Network

The use of UV (ultraviolet fluorescence) light in microscopy allows improving the quality of images and observation of structures that are not visible in visible spectrum. The disadvantage of this method is the degradation of microstructures in the slide due to exposure to UV light. The article examines the possibility of using a convolutional neural network to perform this type of conversion without damaging the slides. Using eosin hematoxylin stained slides, a database of image pairs was created for visible light (halogen lamp) and UV light. This database was used to train a multi–layer unidirectional convolutional neural network. The results of the study were subjectively and objectively assessed using the SSIM (Structural Similarity Index Measure) and SSIM (structure only) image quality measures. The results show that it is possible to perform this type of conversion (the studies used liver slides for 100× magnification), and in some cases there was an additional improvement in image quality.

Farmer’s Credit Rating Model and Application Based on Multilayer Unified Network with Linear Classifier

This article innovatively builds the infrastructure of farmer credit rating index system into a multilevel unidirectional network structure. First, according to the logical structure of the three-level credit rating index system, a four-level unidirectional network is constructed, and the credit rating calculation formulas of all indexes at the four-level network are established. Furthermore, the special cases of the credit rating formula with the first- and second-level farmer credit rating index system are discussed. On this basis, it is extended to a credit rating index system with more than four levels, and the corresponding credit rating formula is established. Finally, the general formula of credit rating formula of the farmer credit rating index system from first level to multilevel is obtained. In order to solve the problem of farmers' credit rating, this paper also designs a linear segmentation classifier to classify the results of multilayer unidirectional network, establishes the rules of farmers' credit rating and the unidirectional network linear segmentation evaluation model of farmers' credit rating, and discusses the properties of bank credit based on farmers’ credit rating. Finally, the model established in this paper is applied to the credit rating of farmers in A County, Guangdong Province in China. When the credit rating of 160 farmers is carried out, the evaluation results are in line with the actual credit rating of farmers in A County, with an accuracy of 100%. This research has the maneuverability to carry on the scientific credit rating to the countryside. This study has important method guidance and operability for rural credit rating.

Verification of a Nondestructive Method for Assessing the Humidity of Saline Brick Walls in Historical Buildings

The paper presents the results of the verification of the neural method for assessing the humidity of saline brick walls. The method was previously developed by the authors and can be useful for the nondestructive assessment of the humidity of walls in historic buildings when destructive intervention during testing is not possible due to conservation restrictions. However, before being implemented in construction practice, this method requires validation by verification on other historic buildings, which to date has not been done. The paper presents the results of such verification, which has never been carried out before, and thus extends the scope of knowledge related to the issue. For experimental verification of the artificial neural network (ANN), the results of moisture tests of two selected historic buildings, other than those used for ANN learning and testing processes, were used. An artificial unidirectional multilayer neural network with backward error propagation and the algorithm for learning conjugate gradient (CG) was found to be useful for this purpose. The obtained satisfactory value of the linear correlation coefficient R of 0.807 and low average absolute error |Δf| of 1.16% confirms this statement. The values of average relative error |RE| of 19.02%, which were obtained in this research, were not very high for an in-situ study. Moreover, the relative error values |RE| were mostly in the range of 15% to 25%.

Research on Short-term Power Load Forecasting Based on Bi-GRU

The precise, accurate and efficient short-term load forecasting can guide the power supply companies to rationally arrange power dispatch plans, help improve the stability of grid operation, and significantly, improving power utilization, thereby optimizing corporate marketing strategies and increasing corporate economic returns. Short-term load forecasting methods based on deep learning have gained greater attention from academia and power companies. Among them, the load forecasting model based on recurrent neural network has gained excellent forecasting results compared with traditional machine learning models. The advantage of the cyclic neural network is that it can extract the degree of relevance of the data in the time dimension, but the unidirectional network only considers the impact of historical data on the current forecast. This paper proposes a load forecasting model based on a bidirectional gated recurrent unit, which further improves the relevance of data. However, it introduces meteorological factors and the influence of holidays to improve the accuracy of forecast results. Taking the power load data of a district of a city in China as an example, the prediction results of this model meet the actual requirements and show better prediction performance over Bi-LSTM, LSTM, GRU, and other models.

Complex dynamics of some models of antimicrobial resistance on complex networks

Studying the spread of epidemics and diseases is a worldwide problem especially during the current time where the whole world is suffering from COVID‐19 pandemic. Antimicrobial resistance (AMR) and waning vaccination are classified as worldwide problems. Both depend on the exposure time to antibiotic and vaccination. Here, a simple model for competition between drug‐resistant and drug‐sensitive bacteria is given. Conditions for local stability are investigated, which agree with observation. Existence of positive solution in the AMR complex networks is proved. Dynamics of the identical AMR models are explored with different topologies of complex networks such as global, star, line, and unidirectional line networks coupled through their susceptible states. Chaotic attractors are shown to exist as the AMR models are located on all these topologies of complex networks. Thus, it is found that the dynamics of the AMR model become more complicated as it is located on either integer‐order or fractional‐order complex networks. Furthermore, a discretized version of the fractional AMR model is presented. Complex dynamics such as existence of Neimark–Sacker, flip bifurcations, coexistence of multiple attractors, homoclinic connections, and multiple closed invariant curves are investigated. Basin sets of attraction are also computed. Finally, the discretized system is located on complex networks with different topologies which show rich variety of complex dynamics. Also, 0–1 test is used to verify the existence of unpredictable dynamics. So, studying the dynamics of AMR models on complex networks is very helpful to understand the mechanism of spread of diseases.

Droop‐based consensus control scheme for economic dispatch in islanded microgrids

In this study, a droop-based consensus control scheme (DCCS) is proposed to realise economic dispatch for islanded microgrids in a fully distributed manner. Different from the existing consensus-based approaches, the proposed DCCS embeds droop control into the consensus algorithm, which utilises the neighbouring incremental cost information and P–f droop control to realise economic dispatch for islanded microgrids. The implementation of the proposed DCCS only needs local measurements and transmitting incremental cost information among neighbouring distributed generations (DGs) through a unidirectional ring communication network (URCN), which does not require any additional conditions (such as leader DG, power mismatch estimation, etc.). Besides, the URCN in the proposed DCCS can efficiently reduce the communication burden compared with the existing consensus-based algorithms implemented in an undirected manner. Furthermore, the theoretical deduction is given to demonstrate the effectiveness of the proposed DCCS. Several simulations show an excellent performance of the proposed DCCS against sudden load and source changes, and the ability to maintain the steady operation even with communication failure.

Framework of IoT Services over Unidirectional Visible Lights Communication Networks

Visible Light Communication (VLC) has been noted as an emerging technology for communications in wireless local area networks. VLC provides some distinctive features over the conventional wireless access technologies, such as Wi-Fi, Bluetooth, or ZigBee. The most prominent feature of VLC is that it can provide more exact location information, since it is based on a particular light. In addition, VLC can reduce the frequency interferences from numerous wireless channels, since it uses a completely different radio frequency channel from the conventional wireless access technologies. Thus, VLC can be used for Internet-of-Things (IoT) services. Nevertheless, up to now, not enough studies on how to provide IoT services over VLC networks have been conducted. In this paper, we propose a framework to provide IoT services in VLC networks. In particular, we will consider the unidirectional VLC network, in which the downlink channel from the VLC transmitter to the VLC receiver is given by using VLC communication, whereas the uplink channel from the VLC receiver to the VLC transmitter is implemented by using another wireless access technology, such as Wi-Fi. This is because most of the VLC receivers, such as mobile phones, cannot support the uplink VLC communication. Based on the framework of IoT services over unidirectional VLC, in this paper, we also propose the VLC–IoT protocol (VIP) which is an application layer protocol for data transport with the session management functionality that can be used to effectively provide IoT services among IoT servers, VLC transmitters and VLC receivers in the networks. The proposed VIP protocol is implemented by using the Cooja simulator. For performance analysis, the proposed scheme is compared with the existing CoAP-based scheme that does not provide the session management. From a variety of simulation experiments, we see that the proposed scheme can provide lower data transmission and handover delays, compared to the existing scheme.

The performance comparison of hybrid WDM/TDM, TDM and WDM PONs with 128 ONUs

Abstract Here, we have simulated different unidirectional passive optical networks (PONs) technologies such as wavelength division multiplexing (WDM), time-division multiplexing (TDM) and hybridPONs with different users for varying fiber length, data rate, continuous wave laser power and number of users. Their performances based on the quality-factor (Q-factor) and bit error rate (BER) using OptiSystem software 7.0 with using an Erbium doped-fiber amplifier (EDFA) were compared. Our model used 16 and 128 users, where the performance of the unidirectional Hybrid WDM/TDM PONs with 4 wavelengths and 128 user systems have better high Q-factor and lower BER, compared to that of 128 user WDM PON and 128 user TDM PON systems.

Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values

Short-term traffic forecasting based on deep learning methods, especially recurrent neural networks (RNN), has received much attention in recent years. However, the potential of RNN-based models in traffic forecasting has not yet been fully exploited in terms of the predictive power of spatial–temporal data and the capability of handling missing data. In this paper, we focus on RNN-based models and attempt to reformulate the way to incorporate RNN and its variants into traffic prediction models. A stacked bidirectional and unidirectional LSTM network architecture (SBU-LSTM) is proposed to assist the design of neural network structures for traffic state forecasting. As a key component of the architecture, the bidirectional LSTM (BDLSM) is exploited to capture the forward and backward temporal dependencies in spatiotemporal data. To deal with missing values in spatial–temporal data, we also propose a data imputation mechanism in the LSTM structure (LSTM-I) by designing an imputation unit to infer missing values and assist traffic prediction. The bidirectional version of LSTM-I is incorporated in the SBU-LSTM architecture. Two real-world network-wide traffic state datasets are used to conduct experiments and published to facilitate further traffic prediction research. The prediction performance of multiple types of multi-layer LSTM or BDLSTM models is evaluated. Experimental results indicate that the proposed SBU-LSTM architecture, especially the two-layer BDLSTM network, can achieve superior performance for the network-wide traffic prediction in both accuracy and robustness. Further, comprehensive comparison results show that the proposed data imputation mechanism in the RNN-based models can achieve outstanding prediction performance when the model’s input data contains different patterns of missing values.

Mechanical properties of human oral mucosa tissues are site dependent: A combined biomechanical, histological and ultrastructural approach.

AimTo investigate load‐deformation properties of Thiel‐embalmed human oral mucosa tissues and to compare three different anatomical regions in terms of mechanical, histological and ultrastructural characteristic with focus on the extracellular matrix.Materials and MethodsThirty specimens from three different regions of the oral cavity: attached gingiva, buccal mucosa and the hard palate were harvested from two Thiel‐embalmed cadavers. Mechanical properties were obtained, combining strain evaluation and digital image correlation in a standardised approach. Elastic modulus, tensile strength, strain at maximum load and strain to failure were computed and analysed statistically. Subsamples were also analysed using scanning electron microscopy (SEM) and histological analysis.ResultsThe highest elastic modulus of 37.36 ± 17.4 MPa was found in the attached gingiva group, followed by the hard palate and buccal mucosa. The elastic moduli of attached gingiva differed significantly to the buccal mucosa (p = .01) and hard palate (p = .021). However, there was no difference in the elastic moduli between the buccal mucosa and hard palate (p > .22). The tensile strength of the tissue samples ranged from 1.54 ± 0.5MPa to 3.81 ± 0.9 MPa, with a significant difference between gingiva group and buccal mucosa or hard palate (p = .001). No difference was found in the mean tensile strength between the buccal mucosa and hard palate (p = .92). Ultrastructural imaging yielded a morphological basis for the various mechanical properties found intraorally; the attached gingiva showed unidirectional collagen fibre network whereas the buccal mucosa and hard palate showed multi‐directional network, which was more prone to tension failure and less elasticity.ConclusionThis is the first study assessing the various morphological‐mechanical relationships of intraoral soft tissues, utilising Thiel‐embalmed tissues. The findings of this study suggest that the tissues from different intraoral regions showed various morphological‐mechanical behaviour which was also confirmed under the SEM and in the histological analysis.