Breast cancer remains a significant global health challenge, necessitating effective early detection and prognosis to enhance patient outcomes. Current diagnostic methods, including mammography and MRI, suffer from li...Breast cancer remains a significant global health challenge, necessitating effective early detection and prognosis to enhance patient outcomes. Current diagnostic methods, including mammography and MRI, suffer from limitations such as uncertainty and imprecise data, leading to late-stage diagnoses. To address this, various expert systems have been developed, but many rely on type-1 fuzzy logic and lack mobile-based applications for data collection and feedback to healthcare practitioners. This research investigates the development of an Enhanced Mobile-based Fuzzy Expert system (EMFES) for breast cancer pre-growth prognosis. The study explores the use of type-2 fuzzy logic to enhance accuracy and model uncertainty effectively. Additionally, it evaluates the advantages of employing the python programming language over java for implementation and considers specific risk factors for data collection. The research aims to dynamically generate fuzzy rules, adapting to evolving breast cancer research and patient data. Key research questions focus on the comparative effectiveness of type-2 fuzzy logic, the handling of uncertainty and imprecise data, the integration of mobile-based features, the choice of programming language, and the creation of dynamic fuzzy rules. Furthermore, the study examines the differences between the Mamdani Inference System and the Sugeno Fuzzy Inference method and explores challenges and opportunities in deploying the EMFES on mobile devices. The research identifies a critical gap in existing breast cancer diagnostic systems, emphasizing the need for a comprehensive, mobile-enabled, and adaptable solution by developing an EMFES that leverages Type-2 fuzzy logic, the Sugeno Inference Algorithm, Python Programming, and dynamic fuzzy rule generation. This study seeks to enhance early breast cancer detection and ultimately reduce breast cancer-related mortality.展开更多
Addressing classification and prediction challenges, tree ensemble models have gained significant importance. Boosting ensemble techniques are commonly employed for forecasting Type-II diabetes mellitus. Light Gradien...Addressing classification and prediction challenges, tree ensemble models have gained significant importance. Boosting ensemble techniques are commonly employed for forecasting Type-II diabetes mellitus. Light Gradient Boosting Machine (LightGBM) is a widely used algorithm known for its leaf growth strategy, loss reduction, and enhanced training precision. However, LightGBM is prone to overfitting. In contrast, CatBoost utilizes balanced base predictors known as decision tables, which mitigate overfitting risks and significantly improve testing time efficiency. CatBoost’s algorithm structure counteracts gradient boosting biases and incorporates an overfitting detector to stop training early. This study focuses on developing a hybrid model that combines LightGBM and CatBoost to minimize overfitting and improve accuracy by reducing variance. For the purpose of finding the best hyperparameters to use with the underlying learners, the Bayesian hyperparameter optimization method is used. By fine-tuning the regularization parameter values, the hybrid model effectively reduces variance (overfitting). Comparative evaluation against LightGBM, CatBoost, XGBoost, Decision Tree, Random Forest, AdaBoost, and GBM algorithms demonstrates that the hybrid model has the best F1-score (99.37%), recall (99.25%), and accuracy (99.37%). Consequently, the proposed framework holds promise for early diabetes prediction in the healthcare industry and exhibits potential applicability to other datasets sharing similarities with diabetes.展开更多
Computing,serving as the cornerstone of information processing,plays a pivotal role in the digital service era.The"network"and"computing,"responsible for information transmission and processing res...Computing,serving as the cornerstone of information processing,plays a pivotal role in the digital service era.The"network"and"computing,"responsible for information transmission and processing respectively,traditionally belong to different stakeholders and have evolved separately.However,the recent trend toward the coordination and integration of computing and networks has garnered significant attention from both industry and academia.Concepts such as"computility network"and"computing force network"have emerged,and the International Telecommunication Union-Telecommunication Standardization(ITU-T)has initiated efforts to develop standards for the coordination of networking and computing(CNC),focusing on the architecture and framework.展开更多
Requirement gathering for software development project is the most crucial stage and thus requirement engineering (RE) occupies the chief position in the software development. Countless techniques concerning the RE pr...Requirement gathering for software development project is the most crucial stage and thus requirement engineering (RE) occupies the chief position in the software development. Countless techniques concerning the RE processes exist to make sure the requirements are coherent, compact and complete in all respects. In this way different aspects of RE are dissected and detailed upon. A comparison of RE in Agile and RE in Waterfall is expatiated and on the basis of the literature survey the overall Agile RE process is accumulated. Agile being a technique produces high quality software in relatively less time as compared to the conventional waterfall methodology. The paramount objective of this study is to take lessons from RE that Agile method may consider, if quality being the cardinal concern. The study is patterned on the survey of the previous research reported in the coexisting literature and the practices which are being pursued in the area.展开更多
Background Digital twins are virtual representations of devices and processes that capture the physical properties of the environment and operational algorithms/techniques in the context of medical devices and tech-no...Background Digital twins are virtual representations of devices and processes that capture the physical properties of the environment and operational algorithms/techniques in the context of medical devices and tech-nologies.Digital twins may allow healthcare organizations to determine methods of improving medical processes,enhancing patient experience,lowering operating expenses,and extending the value of care.During the present COVID-19 pandemic,various medical devices,such as X-rays and CT scan machines and processes,are constantly being used to collect and analyze medical images.When collecting and processing an extensive volume of data in the form of images,machines and processes sometimes suffer from system failures,creating critical issues for hospitals and patients.Methods To address this,we introduce a digital-twin-based smart healthcare system in-tegrated with medical devices to collect information regarding the current health condition,configuration,and maintenance history of the device/machine/system.Furthermore,medical images,that is,X-rays,are analyzed by using a deep-learning model to detect the infection of COVID-19.The designed system is based on the cascade recurrent convolution neural network(RCNN)architecture.In this architecture,the detector stages are deeper and more sequentially selective against small and close false positives.This architecture is a multi-stage extension of the RCNN model and sequentially trained using the output of one stage for training the other.At each stage,the bounding boxes are adjusted to locate a suitable value of the nearest false positives during the training of the different stages.In this manner,the arrangement of detectors is adjusted to increase the intersection over union,overcoming the problem of overfitting.We train the model by using X-ray images as the model was previously trained on another dataset.Results The developed system achieves good accuracy during the detection phase of COVID-19.The experimental outcomes reveal the efficiency of the detection architecture,which yields a mean average precision rate of 0.94.展开更多
Diabetes mellitus is a metabolic disease that is ranked among the top 10 causes of death by the world health organization.During the last few years,an alarming increase is observed worldwide with a 70%rise in the dise...Diabetes mellitus is a metabolic disease that is ranked among the top 10 causes of death by the world health organization.During the last few years,an alarming increase is observed worldwide with a 70%rise in the disease since 2000 and an 80%rise in male deaths.If untreated,it results in complications of many vital organs of the human body which may lead to fatality.Early detection of diabetes is a task of significant importance to start timely treatment.This study introduces a methodology for the classification of diabetic and normal people using an ensemble machine learning model and feature fusion of Chi-square and principal component analysis.An ensemble model,logistic tree classifier(LTC),is proposed which incorporates logistic regression and extra tree classifier through a soft voting mechanism.Experiments are also performed using several well-known machine learning algorithms to analyze their performance including logistic regression,extra tree classifier,AdaBoost,Gaussian naive Bayes,decision tree,random forest,and k nearest neighbor.In addition,several experiments are carried out using principal component analysis(PCA)and Chi-square(Chi-2)fea-tures to analyze the influence of feature selection on the performance of machine learning classifiers.Results indicate that Chi-2 features show high performance than both PCA features and original features.However,the highest accuracy is obtained when the proposed ensemble model LTC is used with the proposed fea-ture fusion framework-work which achieves a 0.85 accuracy score which is the highest of the available approaches for diabetes prediction.In addition,the statis-tical T-test proves the statistical significance of the proposed approach over other approaches.展开更多
Transmission of data over the internet has become a critical issue as a result of the advancement in technology, since it is possible for pirates to steal the intellectual property of content owners. This paper presen...Transmission of data over the internet has become a critical issue as a result of the advancement in technology, since it is possible for pirates to steal the intellectual property of content owners. This paper presents a new digital watermarking scheme that combines some operators of the Genetic Algorithm (GA) and the Residue Number (RN) System (RNS) to perform encryption on an image, which is embedded into a cover image for the purposes of watermarking. Thus, an image watermarking scheme uses an encrypted image. The secret image is embedded in decomposed frames of the cover image achieved by applying a three-level Discrete Wavelet Transform (DWT). This is to ensure that the secret information is not exposed even when there is a successful attack on the cover information. Content creators can prove ownership of the multimedia content by unveiling the secret information in a court of law. The proposed scheme was tested with sample data using MATLAB2022 and the results of the simulation show a great deal of imperceptibility and robustness as compared to similar existing schemes.展开更多
Early diagnosis of breast cancer does not only increase the chances of survival but also control the diffusion of cancerous cells in the body.Previously,researchers have developed machine learning algorithms in breast...Early diagnosis of breast cancer does not only increase the chances of survival but also control the diffusion of cancerous cells in the body.Previously,researchers have developed machine learning algorithms in breast cancer diagnosis such as Support Vector Machine,K-Nearest Neighbor,Convolutional Neural Network,K-means,Fuzzy C-means,Neural Network,Principle Component Analysis(PCA)and Naive Bayes.Unfortunately these algorithms fall short in one way or another due to high levels of computational complexities.For instance,support vector machine employs feature elimination scheme for eradicating data ambiguity and detecting tumors at initial stage.However this scheme is expensive in terms of execution time.On its part,k-means algorithm employs Euclidean distance to determine the distance between cluster centers and data points.However this scheme does not guarantee high accuracy when executed in different iterations.Although the K-nearest Neighbor algorithm employs feature reduction,principle component analysis and 10 fold cross validation methods for enhancing classification accuracy,it is not efficient in terms of processing time.On the other hand,fuzzy c-means algorithm employs fuzziness value and termination criteria to determine the execution time on datasets.However,it proves to be extensive in terms of computational time due to several iterations and fuzzy measure calculations involved.Similarly,convolutional neural network employed back propagation and classification method but the scheme proves to be slow due to frequent retraining.In addition,the neural network achieves low accuracy in its predictions.Since all these algorithms seem to be expensive and time consuming,it necessary to integrate quantum computing principles with conventional machine learning algorithms.This is because quantum computing has the potential to accelerate computations by simultaneously carrying out calculation on many inputs.In this paper,a review of the current machine learning algorithms for breast cancer prediction is provided.Based on the observed shortcomings,a quantum machine learning based classifier is recommended.The proposed working mechanisms of this classifier are elaborated towards the end of this paper.展开更多
Traffic signal control(TSC)systems are one essential component in intelligent transport systems.However,relevant studies are usually independent of the urban traffic simulation environment,collaborative TSC algorithms...Traffic signal control(TSC)systems are one essential component in intelligent transport systems.However,relevant studies are usually independent of the urban traffic simulation environment,collaborative TSC algorithms and traffic signal communication.In this paper,we propose(1)an integrated and cooperative Internet-of-Things architecture,namely General City Traffic Computing System(GCTCS),which simultaneously leverages an urban traffic simulation environment,TSC algorithms,and traffic signal communication;and(2)a general multi-agent reinforcement learning algorithm,namely General-MARL,considering cooperation and communication between traffic lights for multi-intersection TSC.In experiments,we demonstrate that the integrated and cooperative architecture of GCTCS is much closer to the real-life traffic environment.The General-MARL increases the average movement speed of vehicles in traffic by 23.2%while decreases the network latency by 11.7%.展开更多
Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware reso...Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.展开更多
In this paper, we explore the ability of a hybrid model integrating Long Short-Term Memory (LSTM) networks and eXtreme Gradient Boosting (XGBoost) to enhance the prediction accuracy of Type II Diabetes Mellitus, which...In this paper, we explore the ability of a hybrid model integrating Long Short-Term Memory (LSTM) networks and eXtreme Gradient Boosting (XGBoost) to enhance the prediction accuracy of Type II Diabetes Mellitus, which is caused by a combination of genetic, behavioral, and environmental factors. Utilizing comprehensive datasets from the Women in Data Science (WiDS) Datathon for the years 2020 and 2021, which provide a wide range of patient information required for reliable prediction. The research employs a novel approach by combining LSTM’s ability to analyze sequential data with XGBoost’s strength in handling structured datasets. To prepare this data for analysis, the methodology includes preparing it and implementing the hybrid model. The LSTM model, which excels at processing sequential data, detects temporal patterns and trends in patient history, while XGBoost, known for its classification effectiveness, converts these patterns into predictive insights. Our results demonstrate that the LSTM-XGBoost model can operate effectively with a prediction accuracy achieving 0.99. This study not only shows the usefulness of the hybrid LSTM-XGBoost model in predicting diabetes but it also provides the path for future research. This progress in machine learning applications represents a significant step forward in healthcare, with the potential to alter the treatment of chronic diseases such as diabetes and lead to better patient outcomes.展开更多
With the development of information technology,the Internet of Things(IoT)has gradually become the third wave of the worldwide information industry revolution after the computer and the Internet.The application of the...With the development of information technology,the Internet of Things(IoT)has gradually become the third wave of the worldwide information industry revolution after the computer and the Internet.The application of the IoT has brought great convenience to people’s production and life.However,the potential information security problems in various IoT applications are gradually exposed and people pay more attention to them.The traditional centralized data storage and management model of the IoT is easy to cause transmission delay,single point of failure,privacy disclosure and other problems,and eventually leads to unpredictable behavior of the system.Blockchain technology can effectively improve the operation and data security status of the IoT.Referring to the storage model of the Fabric blockchain project,this paper designs a data security storage model suitable for the IoT system.The simulation results show that the model is not only effective and extensible,but also can better protect the data security of the Internet of Things.展开更多
In this paper, we propose a technique to locate abnormal growth of cells in breast tissue and suggest further pathological test, when require. We compare normal breast tissue with malignant invasive breast tissue by a...In this paper, we propose a technique to locate abnormal growth of cells in breast tissue and suggest further pathological test, when require. We compare normal breast tissue with malignant invasive breast tissue by a series of image processing steps. Normal ductal epithelial cells and ductal/lobular invasive carcinogenic cells also consider for comparison here in this paper. In fact, features of cancerous breast tissue （invasive） are extracted and analyses with normal breast tissue. We also suggest the breast cancer recognition technique through image processing and prevention by controlling p53 gene mutation to some extent.展开更多
The performance of the transfer alignment has great impact on inertial navigation systems.As the transfer alignment is generally implemented using a filter to compensate the errors,its accuracy,rapidity and anti-distu...The performance of the transfer alignment has great impact on inertial navigation systems.As the transfer alignment is generally implemented using a filter to compensate the errors,its accuracy,rapidity and anti-disturbance capability are key properties to evaluate the filtering process.In terms of the superiority in dealing with the noise,H∞filtering has been used to improve the anti-disturbance capability of the transfer alignment.However,there is still a need to incorporate system uncertainty due to various dynamic conditions.Based on the structural value theory,a robustness stability analysis method has been proposed for the transfer alignment to evaluate the impact of uncertainty on the navigation system.The mathematical derivation has been elaborated in this paper,and the simulation has been carried out to verify the effectiveness of the algorithm.展开更多
Performance degradation or system resource exhaustion can be attributed to inadequate computing resources as a result of software aging.In the real world,the workload of a web server varies with time,which will cause ...Performance degradation or system resource exhaustion can be attributed to inadequate computing resources as a result of software aging.In the real world,the workload of a web server varies with time,which will cause a nonlinear aging phenomenon.The nonlinear property often makes analysis and modelling difficult.Workload is one of the important factors influencing the speed of aging.This paper quantitatively analyzes the workload-aging relation and proposes a framework for aging control under varying workloads.In addition,this paper proposes an approach that employs prior information of workloads to accurately forecast incoming system exhaustion.The workload data are used as a threshold to divide the system resource usage data into multiple sections,while in each section the workload data can be treated as a constant.Each section is described by an individual autoregression(AR)model.Compared with other AR models,the proposed approach can forecast the aging process with a higher accuracy.展开更多
The internal structure of the inertial measurement unit(IMU)in active state is easily damaged in the high overload environment.So that the IMU is usually required to be powered within the disappearance of the high ove...The internal structure of the inertial measurement unit(IMU)in active state is easily damaged in the high overload environment.So that the IMU is usually required to be powered within the disappearance of the high overload.In this paper,a mechanical switch is designed to enable the IMU based on the analysis of the impact of high overload on the power-supply circuit.In which,parameters of mechanical switch are determined through theoretical calculation and data analysis.The innovation of the proposed structure lies in that the mechanical switch is triggered through the high overload process and could provide a delay signal for the circuit.After all,the proposed switch is tested through mechanical simulation,impact test and practical test.The experimental results show that the designed mechanical switch can effectively and reliably provide the delay for the circuit and guarantee operation of the IMU under high overload.展开更多
The wind as a natural phenomenon would cause the derivation of the pesticide drops during the operation of agricultural unmanned aerial vehicles(UAV).In particular,the changeable wind makes it difficult for the precis...The wind as a natural phenomenon would cause the derivation of the pesticide drops during the operation of agricultural unmanned aerial vehicles(UAV).In particular,the changeable wind makes it difficult for the precision agriculture.For accurate spraying of pesticide,it is necessary to estimate the real-time wind parameters to provide the correction reference for the UAV path.Most estimation algorithms are model based,and as such,serious errors can arise when the models fail to properly fit the physical wind motions.To address this problem,a robust estimation model is proposed in this paper.Considering the diversity of the wind,three elemental time-related Markov models with carefully designed parameterαare adopted in the interacting multiple model(IMM)algorithm,to accomplish the estimation of the wind parameters.Furthermore,the estimation accuracy is dependent as well on the filtering technique.In that regard,the sparse grid quadrature Kalman filter(SGQKF)is employed to comprise the computation load and high filtering accuracy.Finally,the proposed algorithm is ran using simulation tests which results demonstrate its effectiveness and superiority in tracking the wind change.展开更多
Malicious traffic detection over the internet is one of the challenging areas for researchers to protect network infrastructures from any malicious activity.Several shortcomings of a network system can be leveraged by...Malicious traffic detection over the internet is one of the challenging areas for researchers to protect network infrastructures from any malicious activity.Several shortcomings of a network system can be leveraged by an attacker to get unauthorized access through malicious traffic.Safeguard from such attacks requires an efficient automatic system that can detect malicious traffic timely and avoid system damage.Currently,many automated systems can detect malicious activity,however,the efficacy and accuracy need further improvement to detect malicious traffic from multi-domain systems.The present study focuses on the detection of malicious traffic with high accuracy using machine learning techniques.The proposed approach used two datasets UNSW-NB15 and IoTID20 which contain the data for IoT-based traffic and local network traffic,respectively.Both datasets were combined to increase the capability of the proposed approach in detecting malicious traffic from local and IoT networks,with high accuracy.Horizontally merging both datasets requires an equal number of features which was achieved by reducing feature count to 30 for each dataset by leveraging principal component analysis(PCA).The proposed model incorporates stacked ensemble model extra boosting forest(EBF)which is a combination of tree-based models such as extra tree classifier,gradient boosting classifier,and random forest using a stacked ensemble approach.Empirical results show that EBF performed significantly better and achieved the highest accuracy score of 0.985 and 0.984 on the multi-domain dataset for two and four classes,respectively.展开更多
Technology trends such as Software-Defined Networking (SDN) are transforming networking services in terms of flexibility and faster deployment times. SDN separates the control plane from the data plane with its centra...Technology trends such as Software-Defined Networking (SDN) are transforming networking services in terms of flexibility and faster deployment times. SDN separates the control plane from the data plane with its centralised architecture compared with the distributed approach used in other management systems. However, management systems are still required to adapt the new emerging SDN-like technologies to address various security and complex management issues. Simple Network Management Protocol (SNMP) is the most widespread management protocol implemented in a traditional Network Management System (NMS) but has some limitations with the development of SDN-like services. Hence, many studies have been undertaken to merge the SDN-like services with traditional network management systems. Results show that merging SDN with traditional NMS systems not only increases the average Management Information Base (MIB) polling time but also creates additional overheads on the network. Therefore, this paper proposes a dynamic scheme for MIB polling using an additional MIB controller agent within the SDN controller. Our results show that using the proposed scheme, the average polling time can be significantly reduced (i.e., faster polling of the MIB information) and also requires very low overhead because of the small sized OpenFlow messages used during polling.展开更多
We present a problem for benchmarking the robustness of cellular up-links, in an automatic weather station (AWS) testbed. Based on the problem, we conduct a small-scale measurement study of robustness, where the AWS i...We present a problem for benchmarking the robustness of cellular up-links, in an automatic weather station (AWS) testbed. Based on the problem, we conduct a small-scale measurement study of robustness, where the AWS is equipped with four (4) cellular modems for weather data delivery. The effectiveness of up-links is challenging because of overlapping spatial-temporal factors such as the presence of good reflectors that lead to multi-path effects, interference, network load or other reasons. We argue that, there is a strong need for independent assessments of their robustness, to perform end-to-end network measurement. However, it is yet difficult to go from a particular measurement to an assessment of the entire network. We extensively measure the variability of Radio Signal Strength (RSSI) as link metric on the cellular modems. The RSSI is one of the important link metrics that can determine the robustness of received RF signals, and explore how they differed from one another at a particular location and instant time. We also apply the statistical analysis that quantifies the level of stability by considering the robustness, referring short-term variation, and determines good up-link in comparison to weak one. The results show that the robustness of cellular up-links exists for an unpredictable period of time and lower than one could hope. More than 50% of up-links are intermittent. Therefore, we plan to extend our work by exploring RSSI thresholds, to develop a classification scheme supporting a decision whether a link is either intermittent or not. This will help in normalizing the level of stability, to design the RSSI estimation metric for the robust routing protocol in weather data networks.展开更多
文摘Breast cancer remains a significant global health challenge, necessitating effective early detection and prognosis to enhance patient outcomes. Current diagnostic methods, including mammography and MRI, suffer from limitations such as uncertainty and imprecise data, leading to late-stage diagnoses. To address this, various expert systems have been developed, but many rely on type-1 fuzzy logic and lack mobile-based applications for data collection and feedback to healthcare practitioners. This research investigates the development of an Enhanced Mobile-based Fuzzy Expert system (EMFES) for breast cancer pre-growth prognosis. The study explores the use of type-2 fuzzy logic to enhance accuracy and model uncertainty effectively. Additionally, it evaluates the advantages of employing the python programming language over java for implementation and considers specific risk factors for data collection. The research aims to dynamically generate fuzzy rules, adapting to evolving breast cancer research and patient data. Key research questions focus on the comparative effectiveness of type-2 fuzzy logic, the handling of uncertainty and imprecise data, the integration of mobile-based features, the choice of programming language, and the creation of dynamic fuzzy rules. Furthermore, the study examines the differences between the Mamdani Inference System and the Sugeno Fuzzy Inference method and explores challenges and opportunities in deploying the EMFES on mobile devices. The research identifies a critical gap in existing breast cancer diagnostic systems, emphasizing the need for a comprehensive, mobile-enabled, and adaptable solution by developing an EMFES that leverages Type-2 fuzzy logic, the Sugeno Inference Algorithm, Python Programming, and dynamic fuzzy rule generation. This study seeks to enhance early breast cancer detection and ultimately reduce breast cancer-related mortality.
文摘Addressing classification and prediction challenges, tree ensemble models have gained significant importance. Boosting ensemble techniques are commonly employed for forecasting Type-II diabetes mellitus. Light Gradient Boosting Machine (LightGBM) is a widely used algorithm known for its leaf growth strategy, loss reduction, and enhanced training precision. However, LightGBM is prone to overfitting. In contrast, CatBoost utilizes balanced base predictors known as decision tables, which mitigate overfitting risks and significantly improve testing time efficiency. CatBoost’s algorithm structure counteracts gradient boosting biases and incorporates an overfitting detector to stop training early. This study focuses on developing a hybrid model that combines LightGBM and CatBoost to minimize overfitting and improve accuracy by reducing variance. For the purpose of finding the best hyperparameters to use with the underlying learners, the Bayesian hyperparameter optimization method is used. By fine-tuning the regularization parameter values, the hybrid model effectively reduces variance (overfitting). Comparative evaluation against LightGBM, CatBoost, XGBoost, Decision Tree, Random Forest, AdaBoost, and GBM algorithms demonstrates that the hybrid model has the best F1-score (99.37%), recall (99.25%), and accuracy (99.37%). Consequently, the proposed framework holds promise for early diabetes prediction in the healthcare industry and exhibits potential applicability to other datasets sharing similarities with diabetes.
文摘Computing,serving as the cornerstone of information processing,plays a pivotal role in the digital service era.The"network"and"computing,"responsible for information transmission and processing respectively,traditionally belong to different stakeholders and have evolved separately.However,the recent trend toward the coordination and integration of computing and networks has garnered significant attention from both industry and academia.Concepts such as"computility network"and"computing force network"have emerged,and the International Telecommunication Union-Telecommunication Standardization(ITU-T)has initiated efforts to develop standards for the coordination of networking and computing(CNC),focusing on the architecture and framework.
文摘Requirement gathering for software development project is the most crucial stage and thus requirement engineering (RE) occupies the chief position in the software development. Countless techniques concerning the RE processes exist to make sure the requirements are coherent, compact and complete in all respects. In this way different aspects of RE are dissected and detailed upon. A comparison of RE in Agile and RE in Waterfall is expatiated and on the basis of the literature survey the overall Agile RE process is accumulated. Agile being a technique produces high quality software in relatively less time as compared to the conventional waterfall methodology. The paramount objective of this study is to take lessons from RE that Agile method may consider, if quality being the cardinal concern. The study is patterned on the survey of the previous research reported in the coexisting literature and the practices which are being pursued in the area.
文摘Background Digital twins are virtual representations of devices and processes that capture the physical properties of the environment and operational algorithms/techniques in the context of medical devices and tech-nologies.Digital twins may allow healthcare organizations to determine methods of improving medical processes,enhancing patient experience,lowering operating expenses,and extending the value of care.During the present COVID-19 pandemic,various medical devices,such as X-rays and CT scan machines and processes,are constantly being used to collect and analyze medical images.When collecting and processing an extensive volume of data in the form of images,machines and processes sometimes suffer from system failures,creating critical issues for hospitals and patients.Methods To address this,we introduce a digital-twin-based smart healthcare system in-tegrated with medical devices to collect information regarding the current health condition,configuration,and maintenance history of the device/machine/system.Furthermore,medical images,that is,X-rays,are analyzed by using a deep-learning model to detect the infection of COVID-19.The designed system is based on the cascade recurrent convolution neural network(RCNN)architecture.In this architecture,the detector stages are deeper and more sequentially selective against small and close false positives.This architecture is a multi-stage extension of the RCNN model and sequentially trained using the output of one stage for training the other.At each stage,the bounding boxes are adjusted to locate a suitable value of the nearest false positives during the training of the different stages.In this manner,the arrangement of detectors is adjusted to increase the intersection over union,overcoming the problem of overfitting.We train the model by using X-ray images as the model was previously trained on another dataset.Results The developed system achieves good accuracy during the detection phase of COVID-19.The experimental outcomes reveal the efficiency of the detection architecture,which yields a mean average precision rate of 0.94.
基金supported by the Florida Center for Advanced Analytics and Data Science funded by Ernesto.Net(under the Algorithms for Good Grant).
文摘Diabetes mellitus is a metabolic disease that is ranked among the top 10 causes of death by the world health organization.During the last few years,an alarming increase is observed worldwide with a 70%rise in the disease since 2000 and an 80%rise in male deaths.If untreated,it results in complications of many vital organs of the human body which may lead to fatality.Early detection of diabetes is a task of significant importance to start timely treatment.This study introduces a methodology for the classification of diabetic and normal people using an ensemble machine learning model and feature fusion of Chi-square and principal component analysis.An ensemble model,logistic tree classifier(LTC),is proposed which incorporates logistic regression and extra tree classifier through a soft voting mechanism.Experiments are also performed using several well-known machine learning algorithms to analyze their performance including logistic regression,extra tree classifier,AdaBoost,Gaussian naive Bayes,decision tree,random forest,and k nearest neighbor.In addition,several experiments are carried out using principal component analysis(PCA)and Chi-square(Chi-2)fea-tures to analyze the influence of feature selection on the performance of machine learning classifiers.Results indicate that Chi-2 features show high performance than both PCA features and original features.However,the highest accuracy is obtained when the proposed ensemble model LTC is used with the proposed fea-ture fusion framework-work which achieves a 0.85 accuracy score which is the highest of the available approaches for diabetes prediction.In addition,the statis-tical T-test proves the statistical significance of the proposed approach over other approaches.
文摘Transmission of data over the internet has become a critical issue as a result of the advancement in technology, since it is possible for pirates to steal the intellectual property of content owners. This paper presents a new digital watermarking scheme that combines some operators of the Genetic Algorithm (GA) and the Residue Number (RN) System (RNS) to perform encryption on an image, which is embedded into a cover image for the purposes of watermarking. Thus, an image watermarking scheme uses an encrypted image. The secret image is embedded in decomposed frames of the cover image achieved by applying a three-level Discrete Wavelet Transform (DWT). This is to ensure that the secret information is not exposed even when there is a successful attack on the cover information. Content creators can prove ownership of the multimedia content by unveiling the secret information in a court of law. The proposed scheme was tested with sample data using MATLAB2022 and the results of the simulation show a great deal of imperceptibility and robustness as compared to similar existing schemes.
文摘Early diagnosis of breast cancer does not only increase the chances of survival but also control the diffusion of cancerous cells in the body.Previously,researchers have developed machine learning algorithms in breast cancer diagnosis such as Support Vector Machine,K-Nearest Neighbor,Convolutional Neural Network,K-means,Fuzzy C-means,Neural Network,Principle Component Analysis(PCA)and Naive Bayes.Unfortunately these algorithms fall short in one way or another due to high levels of computational complexities.For instance,support vector machine employs feature elimination scheme for eradicating data ambiguity and detecting tumors at initial stage.However this scheme is expensive in terms of execution time.On its part,k-means algorithm employs Euclidean distance to determine the distance between cluster centers and data points.However this scheme does not guarantee high accuracy when executed in different iterations.Although the K-nearest Neighbor algorithm employs feature reduction,principle component analysis and 10 fold cross validation methods for enhancing classification accuracy,it is not efficient in terms of processing time.On the other hand,fuzzy c-means algorithm employs fuzziness value and termination criteria to determine the execution time on datasets.However,it proves to be extensive in terms of computational time due to several iterations and fuzzy measure calculations involved.Similarly,convolutional neural network employed back propagation and classification method but the scheme proves to be slow due to frequent retraining.In addition,the neural network achieves low accuracy in its predictions.Since all these algorithms seem to be expensive and time consuming,it necessary to integrate quantum computing principles with conventional machine learning algorithms.This is because quantum computing has the potential to accelerate computations by simultaneously carrying out calculation on many inputs.In this paper,a review of the current machine learning algorithms for breast cancer prediction is provided.Based on the observed shortcomings,a quantum machine learning based classifier is recommended.The proposed working mechanisms of this classifier are elaborated towards the end of this paper.
基金supported by the National Natural Science Foundation of China(Grant Nos.61673150,11622538).
文摘Traffic signal control(TSC)systems are one essential component in intelligent transport systems.However,relevant studies are usually independent of the urban traffic simulation environment,collaborative TSC algorithms and traffic signal communication.In this paper,we propose(1)an integrated and cooperative Internet-of-Things architecture,namely General City Traffic Computing System(GCTCS),which simultaneously leverages an urban traffic simulation environment,TSC algorithms,and traffic signal communication;and(2)a general multi-agent reinforcement learning algorithm,namely General-MARL,considering cooperation and communication between traffic lights for multi-intersection TSC.In experiments,we demonstrate that the integrated and cooperative architecture of GCTCS is much closer to the real-life traffic environment.The General-MARL increases the average movement speed of vehicles in traffic by 23.2%while decreases the network latency by 11.7%.
文摘Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.
文摘In this paper, we explore the ability of a hybrid model integrating Long Short-Term Memory (LSTM) networks and eXtreme Gradient Boosting (XGBoost) to enhance the prediction accuracy of Type II Diabetes Mellitus, which is caused by a combination of genetic, behavioral, and environmental factors. Utilizing comprehensive datasets from the Women in Data Science (WiDS) Datathon for the years 2020 and 2021, which provide a wide range of patient information required for reliable prediction. The research employs a novel approach by combining LSTM’s ability to analyze sequential data with XGBoost’s strength in handling structured datasets. To prepare this data for analysis, the methodology includes preparing it and implementing the hybrid model. The LSTM model, which excels at processing sequential data, detects temporal patterns and trends in patient history, while XGBoost, known for its classification effectiveness, converts these patterns into predictive insights. Our results demonstrate that the LSTM-XGBoost model can operate effectively with a prediction accuracy achieving 0.99. This study not only shows the usefulness of the hybrid LSTM-XGBoost model in predicting diabetes but it also provides the path for future research. This progress in machine learning applications represents a significant step forward in healthcare, with the potential to alter the treatment of chronic diseases such as diabetes and lead to better patient outcomes.
基金supported by the National Social Science Foundation Project of China under Grant 16BTQ085.
文摘With the development of information technology,the Internet of Things(IoT)has gradually become the third wave of the worldwide information industry revolution after the computer and the Internet.The application of the IoT has brought great convenience to people’s production and life.However,the potential information security problems in various IoT applications are gradually exposed and people pay more attention to them.The traditional centralized data storage and management model of the IoT is easy to cause transmission delay,single point of failure,privacy disclosure and other problems,and eventually leads to unpredictable behavior of the system.Blockchain technology can effectively improve the operation and data security status of the IoT.Referring to the storage model of the Fabric blockchain project,this paper designs a data security storage model suitable for the IoT system.The simulation results show that the model is not only effective and extensible,but also can better protect the data security of the Internet of Things.
文摘In this paper, we propose a technique to locate abnormal growth of cells in breast tissue and suggest further pathological test, when require. We compare normal breast tissue with malignant invasive breast tissue by a series of image processing steps. Normal ductal epithelial cells and ductal/lobular invasive carcinogenic cells also consider for comparison here in this paper. In fact, features of cancerous breast tissue （invasive） are extracted and analyses with normal breast tissue. We also suggest the breast cancer recognition technique through image processing and prevention by controlling p53 gene mutation to some extent.
基金This work is supported by National Natural Science Foundation of China,No.61803203and the Fundamental Research Funds for the Central Universities,No.30918011305.
文摘The performance of the transfer alignment has great impact on inertial navigation systems.As the transfer alignment is generally implemented using a filter to compensate the errors,its accuracy,rapidity and anti-disturbance capability are key properties to evaluate the filtering process.In terms of the superiority in dealing with the noise,H∞filtering has been used to improve the anti-disturbance capability of the transfer alignment.However,there is still a need to incorporate system uncertainty due to various dynamic conditions.Based on the structural value theory,a robustness stability analysis method has been proposed for the transfer alignment to evaluate the impact of uncertainty on the navigation system.The mathematical derivation has been elaborated in this paper,and the simulation has been carried out to verify the effectiveness of the algorithm.
基金supported by the Natural Science Foundation of Tianjin(19JCYBJC15900)the National Key Research and Development Program of China(2018YFC0823701)+1 种基金an Open Fund of Tianjin Key Lab for Advanced Signal Processing(2017ASP-TJ04)a linkage grant of the Australian Research Council(LP160101691)
文摘Performance degradation or system resource exhaustion can be attributed to inadequate computing resources as a result of software aging.In the real world,the workload of a web server varies with time,which will cause a nonlinear aging phenomenon.The nonlinear property often makes analysis and modelling difficult.Workload is one of the important factors influencing the speed of aging.This paper quantitatively analyzes the workload-aging relation and proposes a framework for aging control under varying workloads.In addition,this paper proposes an approach that employs prior information of workloads to accurately forecast incoming system exhaustion.The workload data are used as a threshold to divide the system resource usage data into multiple sections,while in each section the workload data can be treated as a constant.Each section is described by an individual autoregression(AR)model.Compared with other AR models,the proposed approach can forecast the aging process with a higher accuracy.
基金This work was supported by the National Natural Science Foundation of China(No.61803203).
文摘The internal structure of the inertial measurement unit(IMU)in active state is easily damaged in the high overload environment.So that the IMU is usually required to be powered within the disappearance of the high overload.In this paper,a mechanical switch is designed to enable the IMU based on the analysis of the impact of high overload on the power-supply circuit.In which,parameters of mechanical switch are determined through theoretical calculation and data analysis.The innovation of the proposed structure lies in that the mechanical switch is triggered through the high overload process and could provide a delay signal for the circuit.After all,the proposed switch is tested through mechanical simulation,impact test and practical test.The experimental results show that the designed mechanical switch can effectively and reliably provide the delay for the circuit and guarantee operation of the IMU under high overload.
基金This work was supported by the National Natural Science Foundation of China(No.61803203).
文摘The wind as a natural phenomenon would cause the derivation of the pesticide drops during the operation of agricultural unmanned aerial vehicles(UAV).In particular,the changeable wind makes it difficult for the precision agriculture.For accurate spraying of pesticide,it is necessary to estimate the real-time wind parameters to provide the correction reference for the UAV path.Most estimation algorithms are model based,and as such,serious errors can arise when the models fail to properly fit the physical wind motions.To address this problem,a robust estimation model is proposed in this paper.Considering the diversity of the wind,three elemental time-related Markov models with carefully designed parameterαare adopted in the interacting multiple model(IMM)algorithm,to accomplish the estimation of the wind parameters.Furthermore,the estimation accuracy is dependent as well on the filtering technique.In that regard,the sparse grid quadrature Kalman filter(SGQKF)is employed to comprise the computation load and high filtering accuracy.Finally,the proposed algorithm is ran using simulation tests which results demonstrate its effectiveness and superiority in tracking the wind change.
文摘Malicious traffic detection over the internet is one of the challenging areas for researchers to protect network infrastructures from any malicious activity.Several shortcomings of a network system can be leveraged by an attacker to get unauthorized access through malicious traffic.Safeguard from such attacks requires an efficient automatic system that can detect malicious traffic timely and avoid system damage.Currently,many automated systems can detect malicious activity,however,the efficacy and accuracy need further improvement to detect malicious traffic from multi-domain systems.The present study focuses on the detection of malicious traffic with high accuracy using machine learning techniques.The proposed approach used two datasets UNSW-NB15 and IoTID20 which contain the data for IoT-based traffic and local network traffic,respectively.Both datasets were combined to increase the capability of the proposed approach in detecting malicious traffic from local and IoT networks,with high accuracy.Horizontally merging both datasets requires an equal number of features which was achieved by reducing feature count to 30 for each dataset by leveraging principal component analysis(PCA).The proposed model incorporates stacked ensemble model extra boosting forest(EBF)which is a combination of tree-based models such as extra tree classifier,gradient boosting classifier,and random forest using a stacked ensemble approach.Empirical results show that EBF performed significantly better and achieved the highest accuracy score of 0.985 and 0.984 on the multi-domain dataset for two and four classes,respectively.
文摘Technology trends such as Software-Defined Networking (SDN) are transforming networking services in terms of flexibility and faster deployment times. SDN separates the control plane from the data plane with its centralised architecture compared with the distributed approach used in other management systems. However, management systems are still required to adapt the new emerging SDN-like technologies to address various security and complex management issues. Simple Network Management Protocol (SNMP) is the most widespread management protocol implemented in a traditional Network Management System (NMS) but has some limitations with the development of SDN-like services. Hence, many studies have been undertaken to merge the SDN-like services with traditional network management systems. Results show that merging SDN with traditional NMS systems not only increases the average Management Information Base (MIB) polling time but also creates additional overheads on the network. Therefore, this paper proposes a dynamic scheme for MIB polling using an additional MIB controller agent within the SDN controller. Our results show that using the proposed scheme, the average polling time can be significantly reduced (i.e., faster polling of the MIB information) and also requires very low overhead because of the small sized OpenFlow messages used during polling.
文摘We present a problem for benchmarking the robustness of cellular up-links, in an automatic weather station (AWS) testbed. Based on the problem, we conduct a small-scale measurement study of robustness, where the AWS is equipped with four (4) cellular modems for weather data delivery. The effectiveness of up-links is challenging because of overlapping spatial-temporal factors such as the presence of good reflectors that lead to multi-path effects, interference, network load or other reasons. We argue that, there is a strong need for independent assessments of their robustness, to perform end-to-end network measurement. However, it is yet difficult to go from a particular measurement to an assessment of the entire network. We extensively measure the variability of Radio Signal Strength (RSSI) as link metric on the cellular modems. The RSSI is one of the important link metrics that can determine the robustness of received RF signals, and explore how they differed from one another at a particular location and instant time. We also apply the statistical analysis that quantifies the level of stability by considering the robustness, referring short-term variation, and determines good up-link in comparison to weak one. The results show that the robustness of cellular up-links exists for an unpredictable period of time and lower than one could hope. More than 50% of up-links are intermittent. Therefore, we plan to extend our work by exploring RSSI thresholds, to develop a classification scheme supporting a decision whether a link is either intermittent or not. This will help in normalizing the level of stability, to design the RSSI estimation metric for the robust routing protocol in weather data networks.
