Uneven Data Research Articles

Bilateral extended FDK: An improved weighting method for static CT imaging.

Improving imaging speed has always been the focus of research in CT technology, which is related to the radiation dose and imaging quality of moving organs, including heart and blood vessels. However, it is difficult to achieve further improvement by increasing the rotation speed of the gantry due to its structural strength limitation. Differing from the conventional CTs, the static CT employs dozens of ray sources to acquire projection data from different angular ranges, and each source only needs to be rotated in a small range to finish a full 360° scan, thus greatly increasing the scanningspeed. As sources of static CT need to be evenly distributed over 360°, the sources and detectors have to be arranged on two parallel rings independently. Such a geometry can be considered as a special case of CT systems with a significantly large cone angle, that is, a part of the detector is missing in the vicinity of the mid-plane. Due to restriction of upper and lower bounds of the cone angle of the static CT, there are uneven projection data varying in each portion of the reconstruction volume, the conventional analytical or iterative reconstruction methods may introduce artifacts in the reconstructedoutcomes. Following the weighting approach extended FDK (xFDK) by Grimmer etal., we propose an improved bilateral xFDK algorithm (bixFDK), which focuses on the reconstruction of the expanded volume. With the same philosophy as xFDK in terms of weighting function design, bixFDK takes the longitudinal offset of the detector with respect to the source into consideration, making our method applicable to a wide range of CT geometries, especially for the static CT. Based on the proposed bixFDK, a new iterative scheme bixFDK-IR is also constructed to extend the applications to a wide range of scan protocols such as sparse-viewscan. The proposed method has been validated with the simulated phantom data and the actual clinical data of the static CT, and demonstrates that it can ensure good image quality and enlarge the reconstruction volume in z-direction of the staticCT. The bixFDK algorithm is an ideal reconstruction approach for static CT geometry, and the iterative scheme of bixFDK-IR is applicable to a wide range of CT geometries and scan protocols, thus providing a wide range of applicationscenarios.

PointCA: Evaluating the Robustness of 3D Point Cloud Completion Models against Adversarial Examples

Point cloud completion, as the upstream procedure of 3D recognition and segmentation, has become an essential part of many tasks such as navigation and scene understanding. While various point cloud completion models have demonstrated their powerful capabilities, their robustness against adversarial attacks, which have been proven to be fatally malicious towards deep neural networks, remains unknown. In addition, existing attack approaches towards point cloud classifiers cannot be applied to the completion models due to different output forms and attack purposes. In order to evaluate the robustness of the completion models, we propose PointCA, the first adversarial attack against 3D point cloud completion models. PointCA can generate adversarial point clouds that maintain high similarity with the original ones, while being completed as another object with totally different semantic information. Specifically, we minimize the representation discrepancy between the adversarial example and the target point set to jointly explore the adversarial point clouds in the geometry space and the feature space. Furthermore, to launch a stealthier attack, we innovatively employ the neighbourhood density information to tailor the perturbation constraint, leading to geometry-aware and distribution-adaptive modifications for each point. Extensive experiments against different premier point cloud completion networks show that PointCA can cause the performance degradation from 77.9% to 16.7%, with the structure chamfer distance kept below 0.01. We conclude that existing completion models are severely vulnerable to adversarial examples, and state-of-the-art defenses for point cloud classification will be partially invalid when applied to incomplete and uneven point cloud data.

A multi-dimensional trust model for misbehavior detection in vehicular ad hoc networks

Communication security is one of the focus issues of vehicular ad hoc networks (VANETs). Trust model is commonly used to identify malicious attackers or filter false messages in the network, which effectively guarantees secure communication between vehicles. Constrained by the distributed network architecture of VANETs, trust evaluation results may be biased due to uneven data distribution and lack of reference. By colluding with the partners, malicious vehicle launches imbalance attack and zigzag attack to bypass the detection of the trust model. To address these new intelligent attacks, we propose a multi-dimensional trust model (MDT) in this paper, which makes a combined evaluation of different trust attributes from vehicles. In MDT, data collection and trust calculation are deployed in vehicles and trust authority (TA), respectively. Since the model dynamically adjusts the corresponding weight coefficient according to the data fluctuations of each trust indicator by adopting entropy weight method and filters anomalous evaluation results by using the median absolute deviation (MAD) algorithm, it can effectively deal with complex and variable intelligent attacks. The legitimate identity of a vehicle will be revoked from the network by TA once its trust value is less than the threshold. We have conducted extensive experimental studies to evaluate the performance of MDT. The experimental results show that the MDT model can accurately and effectively calculate the global trust of vehicles and detect malicious attacks in the network under different scenarios and tests.

E-ReInForMIF Routing Algorithm Based on Energy Selection and Erasure Code Tolerance Machine

Aiming at the problems of data loss and uneven energy consumption in wireless sensor networks during data transmission, this paper proposes a ReInForM transmission fault-tolerant routing algorithm based on energy selection and erasure code fault-tolerant machines (E-ReInForMIF). The E-ReInForMIF algorithm improves the multi-path routing algorithm by combining an erasure coding fault-tolerant machine and node residual energy sorting selection. First, the erasure coding fault-tolerant machine is used to encode the signal, determine the number of transmission paths through multi-path routing, and then select the specific node of the next hop by sorting the residual energy of the node. The E-ReInForMIF routing algorithm effectively solves the problems of uneven energy consumption and data loss in data transmission, improving network lifespan and transmission reliability. Finally, the signal is decoded. The simulation results show that the E-ReInForMIF routing algorithm is superior to the ReInForM routing algorithm in improving the reliability of data transmission.

Effect of Alkaline Treatment on Mechanical and Thermal Properties of Miswak (Salvadora persica) Fiber-Reinforced Polylactic Acid.

This study examines the effects of alkaline treatment on the mechanical and thermal properties of miswak fiber-reinforced polylactic acid. The treatment was performed with three distinct concentrations of sodium hydroxide (NaOH): 1 wt %, 2 wt %, and 3 wt %. The difficulties of interaction between the surface of the fiber and the matrix, which led to this treatment, is caused by miswak fiber's hydrophilic character, which impedes its ability to bind with hydrophobic polylactic acid. FTIR, tensile, TGA, and DMA measurements were used to characterize the composite samples. A scanning electron microscope (SEM) was used to examine the microstructures of many broken samples. The treatment is not yet especially effective in enhancing interfacial bonding, as seen by the uneven tensile strength data. The effect of the treated fiber surface significantly improves the tensile strength of miswak fiber-reinforced PLA composites. Tensile strength improves by 18.01%, 6.48%, and 14.50%, respectively, for 1 wt %, 2 wt %, and 3 wt %. Only 2 wt %-treated fiber exhibits an increase of 0.7% in tensile modulus. The modulus decreases by 4.15 % at 1 wt % and by 19.7% at 3 wt %, respectively. The TGA curve for alkali-treated fiber composites demonstrates a slight increase in thermal stability when compared to untreated fiber composites at high temperatures. For DMA, the composites with surface treatment have higher storage moduli than the composite with untreated miswak fiber, especially for the PLA reinforced with 2 wt % alkali miswak fiber, proving the effectiveness of the treatment.

Improvement of Auxiliary Diagnosis of Diabetic Cardiovascular Disease Based on Data Oversampling and Deep Learning

Diabetic cardiovascular disease is a common complication of diabetes, which can lead to high-mortality diseases such as diabetic cardiomyopathy and atherosclerosis in serious cases. Therefore, effective prevention and management of diabetic cardiovascular disease is demanded. Clinical medical data officers are faced with a situation of a small amount of data and uneven data distribution. In this paper, we propose data oversampling synthesis techniques based on weight and extension algorithms. It can combine 1D-convolutional neural networks and long short-term memory neural networks to solve the problem of a lack of original data. First of all, a few samples based on feature weight are synthesized to make the original unbalanced data evenly distributed. Secondly, the original data are extended and corrected to expand the number of samples. Finally, the deep learning algorithm is used to extract features and classify whether the data have diabetic cardiovascular disease. Data synthesis based on weight and extension algorithms was evaluated on the actual medical datasets and obtained an accuracy of 93.53% and specificity of 94.37%, which confirms that it is an improved solution compared to the other algorithms. Hence, this paper contributes not only a substantial saving of human resources but also improves the efficiency of the clinical diagnosis of diabetic cardiovascular disease, which is conducive to the early detection and treatment of diseases.

Secure authentication and privacy-preserving blockchain for industrial internet of things

Blockchain (BC) technology has overtaken Industrial Internet of Things (IIoT) platforms. It is necessary to explore efficient implementation. Fault tolerance, decentralised control, authentication, cryptographic security, immutability, data integrity, and BC smart contracts are recommended IIoT features. If entities are authenticated and trusted, the internet can be used for industrial activities. Despite several methods, communication is insecure due to scalability, dependability, latency, insufficient transmission security, and uneven data loads. The paper created safe User authentication and optimal BC node selection using AFHENN (Fully Homomorphic encryption neural network) for IIoT to solve the problem. Mutual authentication, secrecy, and integrity protect user data. A registration process secures new User authentication. To protect registered data, it uses cryptographic methods like Transient key congruential generator based Elliptic curve cryptography (TKCG-ECC) and Dual keyed Cipolla's Extended Euclidean Algorithm based lattice cryptosystem (DKCEED-LC). To access BCN, the gateway verifies registered users utilising keyed-based Zero Knowledge of Proof (k-ZKP) and Approximation Fully Homomorphic encryption neural network-based Blockchain. Finally, Approximation Fully Homomorphic encryption neural network-based Blockchain networking authenticates data (AFHENN-BCN). The BCN avoids legal selection of miner nodes and harmful activities. Compared to top techniques, the proposed work achieves improved throughput and PDR (Packet Delivery Ratio) values with minimal computing time and strong security.

Urbanization associated changes in biogeochemical cycles.

Urbanization associated changes in biogeochemical cycles.

An Image Recognition Method for Urine Sediment Based on Semi-supervised Learning

Because there are many categories, large morphological differences and few labels of urinary sediment components, and uneven data in urine sediment images recognition, the accuracy and recall rate of the existing urine sediment images recognition methods are not ideal. The main purpose of this paper is to improve the accuracy and recall of urine sediment image recognition by proposing a urine sediment image classification method based on semi-supervised learning. This paper proposes a method based on semi-supervised learning to classify urine sediment images. This algorithm designs a re-parameterization network (US-RepNet) for low-resolution urine sediment microscopic images to extract complex features of urine sediment images. The dual attention module is introduced on Us-RepNet to increase the extraction of fine-grained features from urine sediment images. And the cross-entropy loss (C.E. loss) function is optimized to train an unbiased classifier to improve long-tailed distribution image classification. The experimental results show that the accuracy of proposed method can reach 94% with only a small amount of labeled data for 16 types of urine sediment images under long-tail distribution. The algorithm can recognize most types, and reduces the need for labeled information, while achieving excellent recognition and classification performance. Comprehensive analysis shows that it can be used as a practical reference for urine sediment analysis. • Re-parameterization network is used to recognize low-resolution microscopic images. • Re-parameterization network can effectively extract the features of urine sediment images. • CSAM increases the extraction of fine-grained features. • Unbiased classifier is trained under the improved C.E. loss function.

EPT: A data-driven transformer model for earthquake prediction

The leading causes of earthquakes are crustal movements, plate movements, and collisions. In recent years, many researchers in earthquake prediction have been predicting earthquakes from historical seismic data in local areas. This approach ignores the underlying internal patterns of crustal motion, plate movement, and collisions. This paper proposes a purely data-driven deep learning model called EPT. The model uses gated feature extraction blocks (GFEB) to mine potential crustal motion and plate movement patterns from global historical seismic catalog data. It uses them to aid mainshock prediction in each local, provincial region. Experiments show that this approach improves model prediction accuracy by up to 50 percent. We also use multi-headed self-attention for the first time to capture long-term dependencies within regional time series, highlighting links between focal features and compensating for the difficulty of focusing on longer-term information in long-term time series with long short-term memory networks (LSTM). In addition, we also use the gradient harmonization mechanism classification (GHMC) loss function for the first time in earthquake prediction, effectively addressing the problem of uneven data distribution across different earthquake magnitude ranges. Finally, we validated the effectiveness of the EPT model in five provincial datasets in mainland China, and the experimental results all achieved an accuracy of over 90 percent.

EnPlace: Energy-Aware Network Partitioning for Controller Placement in SDN

This paper addresses the energy-aware controller placement problem (CPP) in Software-Defined Networking (SDN), considering the Internet of Things (IoT) flows. CPP involves partitioning the network into multiple subsets of switches with a single controller assigned to each subset. On the other hand, an energy-aware CPP reduces the energy consumption by link deactivation and ensures that each controller is reachable from the associated switches with a minimal set of active links. Existing literature considers static data traffic and out-of-band control plane having dedicated control links. However, the out-of-band control plane increases the infrastructure cost. Moreover, with IoT devices, SDN experiences uneven data traffic volume due to diverse activation models of the IoT devices. Hence, an energy-aware CPP should consider the effects of dynamic data traffic as improper controller placement and unplanned link deactivation cause link congestion and controller overload. In this work, we present an energy-aware controller placement scheme, named EnPlace, considering in-band control plane and IoT traffic. Additionally, we propose an energy-aware route selection scheme for existing flows. EnPlace increases energy savings significantly as compared to the existing works. In particular, for 200 IoT devices, the proposed scheme increases energy savings by 22.74% as compared to GreCo, an existing scheme.

Density peaks clustering algorithm based on fuzzy and weighted shared neighbor for uneven density datasets

Uneven density data refers to data with a certain difference in sample density between clusters. The local density of density peaks clustering algorithm (DPC) does not consider the effect of sample density difference between clusters of uneven density data, which may lead to wrong selection of cluster centers; the algorithm allocation strategy makes it easy to incorrectly allocate samples originally belonging to sparse clusters to dense clusters, which reduces clustering efficiency. In this study, we proposed the density peaks clustering algorithm based on fuzzy and weighted shared neighbor for uneven density datasets (DPC-FWSN). First, a nearest neighbor fuzzy kernel function is obtained by combining K-nearest neighbor and fuzzy neighborhood. Then, local density is redefined by the nearest neighbor fuzzy kernel function. The local density can better characterize the distribution characteristics of the sample by balancing the contribution of sample density in dense and sparse areas, in order to avoid the situation that the sparse cluster does not have a cluster center. Finally, the allocation strategy for weighted shared neighbor similarity is proposed to optimize the sample allocation at the boundary of the sparse cluster. Experiments are performed on IDPC-FA, FKNN-DPC, FNDPC, DPCSA and DPC for uneven density datasets, complex morphologies datasets and real datasets. The clustering results demonstrate that DPC-FWSN effectively handles datasets with uneven density distribution.

Efficient Pavement Crack Detection and Classification Using Custom YOLOv7 Model

It is crucial to detect and classify pavement cracks as part of maintaining road safety. The inspection process for identifying and classifying cracks manually is tedious, time-consuming, and potentially dangerous for inspectors. As a result, an efficient automated approach for detecting road cracks is essential for this development. Numerous issues, such as variations in intensity, uneven data availability, the inefficacy of traditional approaches, and others, make it challenging to accomplish. T his research has been carried out to contribute towards developing an efficient pavement crack detection and classification system . This study uses state of the art deep learning algorithm, customized YOLOv7 model. Data from two sources , RDD2022, a publicly available online dataset, and the second set of data gathered from the roads of Malaysia have been used in this investigation. In order to have balanced data for training, many image preprocessing techniques have been applied to the data, such as augmentations, scaling, blurring, etc. Experimental results demonstrate that the detection accuracy of the YOLOv7 model is significant, 92% on the RDD2022 dataset and 88% on our custom dataset. This study reports the outcomes of experiments conducted on both datasets. RDD2022 achieved a precision of 0.9523 and a recall of 0.9545. On the custom dataset, the resulting values for precision and recall were 0.93 and 0.9158, respectively . The results of this study were compared to those of other recent studies in the same field in order to establish a benchmark . Results from the proposed system were more encouraging and surpassed the benchmarking ones.

Load distribution strategy of decentralized databases

Load management in databases typically involves segmentation strategies that determine the distribution and placement of information within data storage systems. This article explores a critical aspect of data segmentation and the challenges associated with selecting the right strategy. Some of the discussed strategies, such as range-based strategy and hashing strategy, come with their limitations and risks that can complicate scalability and data migration operations. The article also highlights issues related to choosing a specific data segmentation pattern in databases, including data movement complexity, uneven load distribution, transactional problems, and data integrity preservation. Choosing the right segmentation strategy is crucial to ensure the efficiency and productivity of a data storage system, and it should be a thoughtful choice tailored to the specific requirements of the system. The article discusses important data segmentation strategies in storage systems, which play a key role in the distribution and management of large datasets. The authors explore three main strategies: the search strategy, the range strategy, and the hashing strategy. Each of these strategies has its unique advantages and characteristics. The search strategy provides maximum control over data placement and the ability to use virtual segments for data balancing. This strategy is particularly useful for multi-user applications but may introduce additional overhead during segment retrieval. The range strategy groups related data into one segment and works well with queries for ranges of values. However, it can lead to uneven load distribution. The hashing strategy helps evenly distribute the load and avoid highly active segments but may introduce additional computational overhead when calculating the hash. The article also discusses data scaling and relocation operations for each of these strategies. For example, the search strategy allows for data scaling and relocation at the user level without system downtime. This article helps readers better understand the differences between data segmentation strategies and choose the most suitable one for their specific use case.

Machine learning in computational chemistry: interplay between (non)linearity, basis sets, and dimensionality.

Machine learning (ML) based methods and tools have now firmly established themselves in physical chemistry and in particular in theoretical and computational chemistry and in materials chemistry. The generality of popular ML techniques such as neural networks or kernel methods (Gaussian process and kernel ridge regression and their flavors) permitted their application to diverse problems from prediction of properties of functional materials (catalysts, solid state ionic conductors, etc.) from descriptors to the building of interatomic potentials (where ML is currently routinely used in applications) and electron density functionals. These ML techniques are assumed to have superior expressive power of nonlinear methods, and are often used "as is", with concepts such as "non-parametric" or "deep learning" used without a clear justification for their need or advantage over simpler and more robust alternatives. In this Perspective, we highlight some interrelations between popular ML techniques and traditional linear regressions and basis expansions and demonstrate that in certain regimes (such as a very high dimensionality) these approximations might collapse. We also discuss ways to recover the expressive power of a nonlinear approach and to help select hyperparameters with the help of high-dimensional model representation and to obtain elements of insight while preserving the generality of the method.

Caring: Towards Collaborative and Cross-domain Wi-Fi Sensing

The quality of a learning-based Wi-Fi sensing system is bounded by the quantity and quality of training data. However, obtaining sufficient and high-quality data across different domains is difficult due to extensive user involvement. We present CARING, a federated-learning-based framework to support collaborative and cross-domain Wi-Fi sensing. A key challenge of CARING is to allow the effective exchange and learning of knowledge across local models that are derived from heterogeneous data sources with uneven data distributions. We overcome this challenge by first extracting the activity-related representation to train local models. The shared global model aggregates received local model parameters and sends them back to individual devices for fine-tuning locally in the deployed environment. By leveraging the crowdsourced knowledge, CARING allows local models to quickly adapt to domain changes using just a few samples seen at test time. We demonstrate the benefit of CARING by applying it to activity recognition across three public datasets collected from 5 environments, 7 deployments, 31 users, and 29 activities. Experimental results show that CARING is highly effective and robust, improving the alternative approach for using single-sourced training data by up to 47%, giving an accuracy of over 80% (up to 100%) for various cross-domain scenarios.

Double-Lumen Endotracheal Tube-Predicting Insertion Depth and Tube Size Based on Patient's Chest X-ray Image Data and 4 Other Body Parameters.

In thoracic surgery, the double lumen endotracheal tube (DLT) is used for differential ventilation of the lung. DLT allows lung collapse on the surgical side that requires access to the thoracic and mediastinal areas. DLT placement for a given patient depends on two settings: a tube of the correct size (or 'size') and to the correct insertion depth (or 'depth'). Incorrect DLT placements cause oxygen desaturation or carbon dioxide retention in the patient, with possible surgical failure. No guideline on these settings is currently available for anesthesiologists, except for the aid by bronchoscopy. In this study, we aimed to predict DLT 'depths' and 'sizes' applied earlier on a group of patients (n = 231) using a computer modeling approach. First, for these patients we retrospectively determined the correlation coefficient (r) of each of the 17 body parameters against 'depth' and 'size'. Those parameters having r > 0.5 and that could be easily obtained or measured were selected. They were, for both DLT settings: (a) sex, (b) height, (c) tracheal diameter (measured from X-ray), and (d) weight. For 'size', a fifth parameter, (e) chest circumference was added. Based on these four or five parameters, we modeled the clinical DLT settings using a Support Vector Machine (SVM). After excluding statistical outliers (±2 SD), 83.5% of the subjects were left for 'depth' in the modeling, and similarly 85.3% for 'size'. SVM predicted 'depths' matched with their clinical values at a r of 0.91, and for 'sizes', at an r of 0.82. The less satisfactory result on 'size' prediction was likely due to the small target choices (n = 4) and the uneven data distribution. Furthermore, SVM outperformed other common models, such as linear regression. In conclusion, this first model for predicting the two DLT key settings gave satisfactory results. Findings would help anesthesiologists in applying DLT procedures more confidently in an evidence-based way.

Dynamic Multichannel Access Based on Deep Reinforcement Learning in Distributed Wireless Networks

With the emergence of innovative applications in vertical industries such as smart home and industrial automation, machine communication has shown a spurt of development. Different from the traditional human-oriented cellular communication, machine communication is characterized by strong uncertainty and abruptness, large-scale concurrent device connection as well as uneven and unsaturated data traffic. This article investigates the dynamic multiple-devices multiple-channels access for unsaturated traffic with retransmission mechanism, which is aimed at reducing the long-term data packet loss resulting from buffer overflows and transmission failure. The instant channel selection will lead to a non-negligible impact on the future decision, motivating us to model this problem as a Markov decision process. Limited by the unknown environment knowledge, we proposed a dynamic access policy based on deep reinforcement learning algorithm to optimally select the channel for transmission or keep silent for Internet-of-Things devices. Simulation results confirm that our proposed channel access strategy can reduce the collision and the packet loss of network. Furthermore, it can also work well when coexisting with the devices that adopt time division multiple access (TDMA) or ALOHA protocol.

A Hybrid Algorithm Incorporating Vector Quantization and One-Class Support Vector Machine for Industrial Anomaly Detection

Anomaly detection plays an important role in industry, especially in ensuring system safety and product quality. Due to the unavailability of anomalous data in many practical cases, anomaly detection is usually solved by one-class classification (OCC) methods using only normal data. As a classical OCC method, one-class support vector machine (OCSVM) is a popular discriminative approach for anomaly detection, which detects abnormal data points by establishing a decision boundary in the kernel space. However, the performance of OCSVM heavily relies on kernel parameters, whose selection is not trivial for anomaly detection problems. Moreover, for some uneven and complex data distributions, different data regions may have quite different densities and shapes, making it difficult for OCSVM to obtain good boundaries in all regions using a global kernel parameter. To address the above two issues, in this article, we propose a hybrid algorithm incorporating vector quantization and OCSVM (VQ-OCSVM). Specifically, vector quantization is used to extract distribution information of normal data, and the results are used to construct an explicit mapping function to map data into a high-dimensional feature space. Then, OCSVM is performed in the feature space to build a classifier. By introducing the explicit mapping into OCSVM, the proposed method can effectively bypass the kernel parameter selection problem of the classical OCSVM method. Furthermore, the constructed mapping carries the data distribution information, and the VQ-OCSVM model can be regarded as an integration of generative learning and discriminative learning. The complementary properties of these two paradigms make the proposed VQ-OCSVM algorithm have better generalization capacity for complex data distribution. Both qualitative and quantitative experimental results demonstrate the effectiveness and advantages of the proposed method.

A Bayesian Approach To Distributed Anomaly Detection In Edge AI Networks

The challenge of anomaly detection is to obtain an accurate understanding of expected behaviour which is intensified when the data are distributed heterogeneously. Transmitting raw data to a central site incurs high communication overhead and raises privacy issues. The concept of Edge AI allows computation to be performed at the edge site allowing for quick decision making in mission critical scenarios such as self-driving cars. A model is learnt locally and its parameters are transmitted and aggregated. However, existing methods of aggregation do not account for variance and heterogeneous distribution of data. They also do not consider edge constraints such as limited computational, memory and communication capabilities of edge devices. In this work, a fully Bayesian approach is employed by means of a Bayesian Random Vector Functional Link AutoEncoder being incorporated with Expectation Propagation for distributed training. Our anomaly detection system operates without any transmission of raw data, is robust under inhomogeneous network densities and under uneven and biased data distributions. It allows for asynchronous updates to converge in a few iterations and is a relatively simple neural network addressing edge constraints without compromising on performance as compared to existing more complex models.