One of the major causes of road accidents is sleepy drivers.Such accidents typically result in fatalities and financial losses and disadvantage other road users.Numerous studies have been conducted to identify the dri...One of the major causes of road accidents is sleepy drivers.Such accidents typically result in fatalities and financial losses and disadvantage other road users.Numerous studies have been conducted to identify the driver’s sleepiness and integrate it into a warning system.Most studies have examined how the mouth and eyelids move.However,this limits the system’s ability to identify drowsiness traits.Therefore,this study designed an Accident Detection Framework(RPK)that could be used to reduce road accidents due to sleepiness and detect the location of accidents.The drowsiness detectionmodel used three facial parameters:Yawning,closed eyes(blinking),and an upright head position.This model used a Convolutional Neural Network(CNN)consisting of two phases.The initial phase involves video processing and facial landmark coordinate detection.The second phase involves developing the extraction of frame-based features using normalization methods.All these phases used OpenCV and TensorFlow.The dataset contained 5017 images with 874 open eyes images,850 closed eyes images,723 open-mouth images,725 closed-mouth images,761 sleepy-head images,and 1084 non-sleepy head images.The dataset of 5017 images was divided into the training set with 4505 images and the testing set with 512 images,with a ratio of 90:10.The results showed that the RPK design could detect sleepiness by using deep learning techniques with high accuracy on all three parameters;namely 98%for eye blinking,96%for mouth yawning,and 97%for head movement.Overall,the test results have provided an overview of how the developed RPK prototype can accurately identify drowsy drivers.These findings will have a significant impact on the improvement of road users’safety and mobility.展开更多
The software engineering field has long focused on creating high-quality software despite limited resources.Detecting defects before the testing stage of software development can enable quality assurance engineers to ...The software engineering field has long focused on creating high-quality software despite limited resources.Detecting defects before the testing stage of software development can enable quality assurance engineers to con-centrate on problematic modules rather than all the modules.This approach can enhance the quality of the final product while lowering development costs.Identifying defective modules early on can allow for early corrections and ensure the timely delivery of a high-quality product that satisfies customers and instills greater confidence in the development team.This process is known as software defect prediction,and it can improve end-product quality while reducing the cost of testing and maintenance.This study proposes a software defect prediction system that utilizes data fusion,feature selection,and ensemble machine learning fusion techniques.A novel filter-based metric selection technique is proposed in the framework to select the optimum features.A three-step nested approach is presented for predicting defective modules to achieve high accuracy.In the first step,three supervised machine learning techniques,including Decision Tree,Support Vector Machines,and Naïve Bayes,are used to detect faulty modules.The second step involves integrating the predictive accuracy of these classification techniques through three ensemble machine-learning methods:Bagging,Voting,and Stacking.Finally,in the third step,a fuzzy logic technique is employed to integrate the predictive accuracy of the ensemble machine learning techniques.The experiments are performed on a fused software defect dataset to ensure that the developed fused ensemble model can perform effectively on diverse datasets.Five NASA datasets are integrated to create the fused dataset:MW1,PC1,PC3,PC4,and CM1.According to the results,the proposed system exhibited superior performance to other advanced techniques for predicting software defects,achieving a remarkable accuracy rate of 92.08%.展开更多
A nursing care planning system that automatically generated nursing summaries from information entered into the Psychiatric Outcome Management System (PSYCHOMS?, Tanioka et al.), was developed to enrich the content of...A nursing care planning system that automatically generated nursing summaries from information entered into the Psychiatric Outcome Management System (PSYCHOMS?, Tanioka et al.), was developed to enrich the content of nursing summaries at psychiatric hospitals, thereby reducing the workload of nurses. Preparing nursing summaries entails finding the required information in nursing records that span a long period of time and then concisely summarizing this information. This time consuming process depends on the clinical experience and writing ability of the nurse. The system described here automatically generates the text data needed for nursing summaries using an algorithm that synthesizes patient information recorded in electronic charts, the Nursing Care Plan information or the data entered for North American Nursing Diagnosis Association (NANDA) 13 domains with predetermined fixed phrases. Advantages of this system are that it enables nursing summaries to be generated automatically in real time, simplifies the process, and permits the standardization of useful nursing summaries that reflect the course of the nursing care provided and its evaluation. Use of this system to automatically generate nursing summaries will allow more nursing time to be devoted to patient care. The system is also useful because it enables nursing summaries that contain the required information to be generated regardless of who prepares them.展开更多
In psychiatric hospitals, the ratios between patients versus physician and patients versus nurse are low as compared to those in general hospitals. Furthermore, usages of electronic medical records are also low so tha...In psychiatric hospitals, the ratios between patients versus physician and patients versus nurse are low as compared to those in general hospitals. Furthermore, usages of electronic medical records are also low so that nurse administrators are limited in their ability to compile, analyze, and generate patient care staffing information for their administrative use. Psychiatric nurse administrators anticipate the development of a nursing administration analysis system that could perform personnel data simulation, manage information on nursing staff, and manage ward/ practice operations. Responding to this situation, the authors developed a nursing administration analysis system utilizing formulae from the Psychiatric Outcome Management System, PSYCHOMS®to aid nurse administrators. Such formulae are awaiting patent approval. The purpose of this study was to examine the validity of the formulae and the Structured Query Language (SQL) statement, and its practical effectiveness of analyzing data. The study findings showed that two kinds of computation expressions—a classification and extraction were able to display required information desired by nurse administrators. Moreover, significant information critical to assigning staff was validated to ensure high quality of nursing care according to the function and characteristic of the hospital ward.展开更多
The combined heat and power economic dispatch(CHPED)problem is a highly intricate energy dispatch challenge that aims to minimize fuel costs while adhering to various constraints.This paper presents a hybrid different...The combined heat and power economic dispatch(CHPED)problem is a highly intricate energy dispatch challenge that aims to minimize fuel costs while adhering to various constraints.This paper presents a hybrid differential evolution(DE)algorithm combined with an improved equilibrium optimizer(DE-IEO)specifically for the CHPED problem.The DE-IEO incorporates three enhancement strategies:a chaotic mechanism for initializing the population,an improved equilibrium pool strategy,and a quasi-opposite based learning mechanism.These strategies enhance the individual utilization capabilities of the equilibrium optimizer,while differential evolution boosts local exploitation and escape capabilities.The IEO enhances global search to enrich the solution space,and DE focuses on local exploitation for more accurate solutions.The effectiveness of DE-IEO is demonstrated through comparative analysis with other metaheuristic optimization algorithms,including PSO,DE,ABC,GWO,WOA,SCA,and equilibrium optimizer(EO).Additionally,improved algorithms such as the enhanced chaotic gray wolf optimization(ACGWO),improved particle swarm with adaptive strategy(MPSO),and enhanced SCA with elite and dynamic opposite learning(EDOLSCA)were tested on the CEC2017 benchmark suite and four CHPED systems with 24,84,96,and 192 units,respectively.The results indicate that the proposed DE-IEO algorithm achieves satisfactory solutions for both the CEC2017 test functions and real-world CHPED optimization problems,offering a viable approach to complex optimization challenges.展开更多
The Cross-domain Heuristic Search Challenge(CHeSC)is a competition focused on creating efficient search algorithms adaptable to diverse problem domains.Selection hyper-heuristics are a class of algorithms that dynamic...The Cross-domain Heuristic Search Challenge(CHeSC)is a competition focused on creating efficient search algorithms adaptable to diverse problem domains.Selection hyper-heuristics are a class of algorithms that dynamically choose heuristics during the search process.Numerous selection hyper-heuristics have different imple-mentation strategies.However,comparisons between them are lacking in the literature,and previous works have not highlighted the beneficial and detrimental implementation methods of different components.The question is how to effectively employ them to produce an efficient search heuristic.Furthermore,the algorithms that competed in the inaugural CHeSC have not been collectively reviewed.This work conducts a review analysis of the top twenty competitors from this competition to identify effective and ineffective strategies influencing algorithmic performance.A summary of the main characteristics and classification of the algorithms is presented.The analysis underlines efficient and inefficient methods in eight key components,including search points,search phases,heuristic selection,move acceptance,feedback,Tabu mechanism,restart mechanism,and low-level heuristic parameter control.This review analyzes the components referencing the competition’s final leaderboard and discusses future research directions for these components.The effective approaches,identified as having the highest quality index,are mixed search point,iterated search phases,relay hybridization selection,threshold acceptance,mixed learning,Tabu heuristics,stochastic restart,and dynamic parameters.Findings are also compared with recent trends in hyper-heuristics.This work enhances the understanding of selection hyper-heuristics,offering valuable insights for researchers and practitioners aiming to develop effective search algorithms for diverse problem domains.展开更多
The forthcoming 6G wireless networks have great potential for establishing AI-based networks that can enhance end-to-end connection and manage massive data of real-time networks.Artificial Intelligence(AI)advancements...The forthcoming 6G wireless networks have great potential for establishing AI-based networks that can enhance end-to-end connection and manage massive data of real-time networks.Artificial Intelligence(AI)advancements have contributed to the development of several innovative technologies by providing sophisticated specific AI mathematical models such as machine learning models,deep learning models,and hybrid models.Furthermore,intelligent resource management allows for self-configuration and autonomous decision-making capabilities of AI methods,which in turn improves the performance of 6G networks.Hence,6G networks rely substantially on AI methods to manage resources.This paper comprehensively surveys the recent work of AI methods-based resource management for 6G networks.Firstly,the AI methods are categorized into Deep Learning(DL),Federated Learning(FL),Reinforcement Learning(RL),and Evolutionary Learning(EL).Then,we analyze the AI approaches according to optimization issues such as user association,channel allocation,power allocation,and mode selection.Thereafter,we provide appropriate solutions to the most significant problems with the existing approaches of AI-based resource management.Finally,various open issues and potential trends related to AI-based resource management applications are presented.In summary,this survey enables researchers to understand these advancements thoroughly and quickly identify remaining challenges that need further investigation.展开更多
The behavioral responses of a tilapia (Oreochromis niloticus) school to low (0.13 mg/L), moderate (0.79 mg/L) and high (2.65 mg/L) levels of unionized ammonia (UIA) concentration were monitored using a computer vision...The behavioral responses of a tilapia (Oreochromis niloticus) school to low (0.13 mg/L), moderate (0.79 mg/L) and high (2.65 mg/L) levels of unionized ammonia (UIA) concentration were monitored using a computer vision system. The swimming activity and geometrical parameters such as location of the gravity center and distribution of the fish school were calculated continuously. These behavioral parameters of tilapia school responded sensitively to moderate and high UIA concen-tration. Under high UIA concentration the fish activity showed a significant increase (P<0.05), exhibiting an avoidance reaction to high ammonia condition, and then decreased gradually. Under moderate and high UIA concentration the school’s vertical location had significantly large fluctuation (P<0.05) with the school moving up to the water surface then down to the bottom of the aquarium alternately and tending to crowd together. After several hours’ exposure to high UIA level, the school finally stayed at the aquarium bottom. These observations indicate that alterations in fish behavior under acute stress can provide important in-formation useful in predicting the stress.展开更多
Three kind of application of ADCP is reported for long-term monitoring in coastal sea.(1)The routine monitoring of water qualities. The water quality and ADCP echo data (600 kHz) observed in the long-term are analgzed...Three kind of application of ADCP is reported for long-term monitoring in coastal sea.(1)The routine monitoring of water qualities. The water quality and ADCP echo data (600 kHz) observed in the long-term are analgzed at MT (Marine Tower) Station of Kansai International Airport in the Osaka Bay, Japan. The correlation between the turbidity and echo intensity in the surface layer is not good because air bubbles generated by breaking wave are not detected by the turbidity meter, but detected well by ADCP. When estimating the turbidity consists of plankton population from echo intensity, the effect of bubbles have to be eliminated. (2) Monitoring stirring up of bottom sediment. The special observation was carried out by using following two ADCP in the Osaka Bay, One ADCP was installed upward on the sea. The other ADCP was hanged downward at the gate type stand about 3 m above from the bottom. At the spring tide, high echo intensities indicating the stirring up of bottom sediment were observed. (3) The monitoring for the boundary condition of water mixing at an estuary. In summer season, the ADCP was set at the mouth of Tanabe Bay in Wakayama Prefecture, Japan. During the observation, water temperature near the bottom showed remarkable falls with interval of about 5-7 d. When the bottom temperature fell, the inflow current with low echo intensity water appears at the bottom layer in the ADCP record. It is concluded that when occasional weak northeast wind makes weak coastal upwelling at the mouth of the bay, the combination of upwelling with internal tidal flow causes remarkable water exchange and dispels the red tide.展开更多
In this article, the authors establish some new nonlinear difference inequalities in two independent variables, which generalize some existing results and can be used as handy tools in the study of qualitative as well...In this article, the authors establish some new nonlinear difference inequalities in two independent variables, which generalize some existing results and can be used as handy tools in the study of qualitative as well as quantitative properties of solutions of certain classes of difference equations.展开更多
Writable XOR executable （W⊕X） and address space layout randomisation （ASLR） have elevated the understanding necessary to perpetrate buffer overflow exploits [1] . However, they have not proved to be a panacea [1 ...Writable XOR executable （W⊕X） and address space layout randomisation （ASLR） have elevated the understanding necessary to perpetrate buffer overflow exploits [1] . However, they have not proved to be a panacea [1 3] , and so other mechanisms, such as stack guards and prelinking, have been introduced. In this paper, we show that host-based protection still does not offer a complete solution. To demonstrate the protection inadequacies, we perform an over the network brute force return-to-libc attack against a preforking concurrent server to gain remote access to a shell. The attack defeats host protection including W⊕X and ASLR. We then demonstrate that deploying a network intrusion detection systems （NIDS） with appropriate signatures can detect this attack efficiently.展开更多
This paper presents a handwritten document recognition system based on the convolutional neural network technique.In today’s world,handwritten document recognition is rapidly attaining the attention of researchers du...This paper presents a handwritten document recognition system based on the convolutional neural network technique.In today’s world,handwritten document recognition is rapidly attaining the attention of researchers due to its promising behavior as assisting technology for visually impaired users.This technology is also helpful for the automatic data entry system.In the proposed systemprepared a dataset of English language handwritten character images.The proposed system has been trained for the large set of sample data and tested on the sample images of user-defined handwritten documents.In this research,multiple experiments get very worthy recognition results.The proposed systemwill first performimage pre-processing stages to prepare data for training using a convolutional neural network.After this processing,the input document is segmented using line,word and character segmentation.The proposed system get the accuracy during the character segmentation up to 86%.Then these segmented characters are sent to a convolutional neural network for their recognition.The recognition and segmentation technique proposed in this paper is providing the most acceptable accurate results on a given dataset.The proposed work approaches to the accuracy of the result during convolutional neural network training up to 93%,and for validation that accuracy slightly decreases with 90.42%.展开更多
One of the most difficult jobs in the post-genomic age is identifying a genetic disease from a massive amount of genetic data.Furthermore,the complicated genetic disease has a very diverse genotype,making it challengi...One of the most difficult jobs in the post-genomic age is identifying a genetic disease from a massive amount of genetic data.Furthermore,the complicated genetic disease has a very diverse genotype,making it challenging to find genetic markers.This is a challenging process since it must be completed effectively and efficiently.This research article focuses largely on which patients are more likely to have a genetic disorder based on numerous medical parameters.Using the patient’s medical history,we used a genetic disease prediction algorithm that predicts if the patient is likely to be diagnosed with a genetic disorder.To predict and categorize the patient with a genetic disease,we utilize several deep and machine learning techniques such as Artificial neural network(ANN),K-nearest neighbors(KNN),and Support vector machine(SVM).To enhance the accuracy of predicting the genetic disease in any patient,a highly efficient approach was utilized to control how the model can be used.To predict genetic disease,deep and machine learning approaches are performed.The most productive tool model provides more precise efficiency.The simulation results demonstrate that by using the proposed model with the ANN,we achieve the highest model performance of 85.7%,84.9%,84.3%accuracy of training,testing and validation respectively.This approach will undoubtedly transform genetic disorder prediction and give a real competitive strategy to save patients’lives.展开更多
This paper presents a Butterfly Optimization Algorithm(BOA)with a wind-driven mechanism for avoiding natural enemies known as WDBOA.To further balance the basic BOA algorithm's exploration and exploitation capabil...This paper presents a Butterfly Optimization Algorithm(BOA)with a wind-driven mechanism for avoiding natural enemies known as WDBOA.To further balance the basic BOA algorithm's exploration and exploitation capabilities,the butterfly actions were divided into downwind and upwind states.The algorithm of exploration ability was improved with the wind,while the algorithm of exploitation ability was improved against the wind.Also,a mechanism of avoiding natural enemies based on Lévy flight was introduced for the purpose of enhancing its global searching ability.Aiming at improving the explorative performance at the initial stages and later stages,the fragrance generation method was modified.To evaluate the effectiveness of the suggested algorithm,a comparative study was done with six classical metaheuristic algorithms and three BOA variant optimization techniques on 18 benchmark functions.Further,the performance of the suggested technique in addressing some complicated problems in various dimensions was evaluated using CEC 2017 and CEC 2020.Finally,the WDBOA algorithm is used proportional-integral-derivative(PID)controller parameter optimization.Experimental results demonstrate that the WDBOA based PID controller has better control performance in comparison with other PID controllers tuned by the Genetic Algorithm(GA),Flower Pollination Algorithm(FPA),Cuckoo Search(CS)and BOA.展开更多
Many organizations apply cloud computing to store and effectively process data for various applications.The user uploads the data in the cloud has less security due to the unreliable verification process of data integ...Many organizations apply cloud computing to store and effectively process data for various applications.The user uploads the data in the cloud has less security due to the unreliable verification process of data integrity.In this research,an enhanced Merkle hash tree method of effective authentication model is proposed in the multi-owner cloud to increase the security of the cloud data.Merkle Hash tree applies the leaf nodes with a hash tag and the non-leaf node contains the table of hash information of child to encrypt the large data.Merkle Hash tree provides the efficient mapping of data and easily identifies the changesmade in the data due to proper structure.The developed model supports privacy-preserving public auditing to provide a secure cloud storage system.The data owners upload the data in the cloud and edit the data using the private key.An enhanced Merkle hash tree method stores the data in the cloud server and splits it into batches.The data files requested by the data owner are audit by a third-party auditor and the multiowner authentication method is applied during the modification process to authenticate the user.The result shows that the proposed method reduces the encryption and decryption time for cloud data storage by 2–167 ms when compared to the existing Advanced Encryption Standard and Blowfish.展开更多
The fifth generation(5G)system is the forthcoming generation of the mobile communication system.It has numerous additional features and offers an extensively high data rate,more capacity,and low latency.However,these ...The fifth generation(5G)system is the forthcoming generation of the mobile communication system.It has numerous additional features and offers an extensively high data rate,more capacity,and low latency.However,these features and applications have many problems and issues in terms of security,which has become a great challenge in the telecommunication industry.This paper aimed to propose a solution to preserve the user identity privacy in the 5G system that can identify permanent identity by using Variable Mobile Subscriber Identity,which randomly changes and does not use the permanent identity between the user equipment and home network.Through this mechanism,the user identity privacy would be secured and hidden.Moreover,it improves the synchronization between mobile users and home networks.Additionally,its compliance with the Authentication and Key Agreement(AKA)structure was adopted in the previous generations.It can be deployed efficiently in the preceding generations because the current architecture imposes minimal modifications on the network parties without changes in the authentication vector’s message size.Moreover,the addition of any hardware to the AKA carries minor adjustments on the network parties.In this paper,the ProVerif is used to verify the proposed scheme.展开更多
Autism spectrum disorder(ASD)is a challenging and complex neurodevelopment syndrome that affects the child’s language,speech,social skills,communication skills,and logical thinking ability.The early detection of ASD ...Autism spectrum disorder(ASD)is a challenging and complex neurodevelopment syndrome that affects the child’s language,speech,social skills,communication skills,and logical thinking ability.The early detection of ASD is essential for delivering effective,timely interventions.Various facial features such as a lack of eye contact,showing uncommon hand or body movements,bab-bling or talking in an unusual tone,and not using common gestures could be used to detect and classify ASD at an early stage.Our study aimed to develop a deep transfer learning model to facilitate the early detection of ASD based on facial fea-tures.A dataset of facial images of autistic and non-autistic children was collected from the Kaggle data repository and was used to develop the transfer learning AlexNet(ASDDTLA)model.Our model achieved a detection accuracy of 87.7%and performed better than other established ASD detection models.Therefore,this model could facilitate the early detection of ASD in clinical practice.展开更多
Escalating cyber security threats and the increased use of Internet of Things(IoT)devices require utilisation of the latest technologies available to supply adequate protection.The aim of Intrusion Detection Systems(I...Escalating cyber security threats and the increased use of Internet of Things(IoT)devices require utilisation of the latest technologies available to supply adequate protection.The aim of Intrusion Detection Systems(IDS)is to prevent malicious attacks that corrupt operations and interrupt data flow,which might have significant impact on critical industries and infrastructure.This research examines existing IDS,based on Artificial Intelligence(AI)for IoT devices,methods,and techniques.The contribution of this study consists of identification of the most effective IDS systems in terms of accuracy,precision,recall and F1-score;this research also considers training time.Results demonstrate that Graph Neural Networks(GNN)have several benefits over other traditional AI frameworks through their ability to achieve in excess of 99%accuracy in a relatively short training time,while also capable of learning from network traffic the inherent characteristics of different cyber-attacks.These findings identify the GNN(a Deep Learning AI method)as the most efficient IDS system.The novelty of this research lies also in the linking between high yielding AI-based IDS algorithms and the AI-based learning approach for data privacy protection.This research recommends Federated Learning(FL)as the AI training model,which increases data privacy protection and reduces network data flow,resulting in a more secure and efficient IDS solution.展开更多
In networking,one major difficulty that nodes suffer from is the need for their addresses to be generated and verified without relying on a third party or public authorized servers.To resolve this issue,the use of sel...In networking,one major difficulty that nodes suffer from is the need for their addresses to be generated and verified without relying on a third party or public authorized servers.To resolve this issue,the use of selfcertifying addresses have become a highly popular and standardized method,of which Cryptographically Generated Addresses(CGA)is a prime example.CGA was primarily designed to deter the theft of IPv6 addresses by binding the generated address to a public key to prove address ownership.Even though the CGA technique is highly effective,this method is still subject to several vulnerabilities with respect to security,in addition to certain limitations in its performance.In this study,the authors present an intensive systematic review of the literature to explore the technical specifications of CGA,its challenges,and existing proposals to enhance the protocol.Given that CGA generation is a time-consuming process,this limitation has hampered the application of CGA in mobile environments where nodes have limited energy and storage.Fulfilling Hash2 conditions in CGA is the heaviest and most timeconsuming part of SEND.To improve the performance of CGA,we replaced the Secure Hash Algorithm(SHA1)with the Message Digest(MD5)hash function.Furthermore,this study also analyzes the possible methods through which a CGA could be attacked.In conducting this analysis,Denial-of-Service(DoS)attacks were identified as the main method of attack toward the CGA verification process,which compromise and threaten the privacy of CGA.Therefore,we propose some modifications to the CGA standard verification algorithm to mitigate DoS attacks and to make CGA more security conscious.展开更多
A novel fiber optic sensor based on hydrogel-immobilized enzyme complex was developed for the simultaneous measurement of dual-parameter,the leap from a single parameter detecting fiber optic sensor to a fiber optic s...A novel fiber optic sensor based on hydrogel-immobilized enzyme complex was developed for the simultaneous measurement of dual-parameter,the leap from a single parameter detecting fiber optic sensor to a fiber optic sensor that can continuously detect two kinds of parameters was achieved.By controlling the temperature from high to low,the function of fiber sulfide sensor and fiber DCP sensor can be realized,so as to realize the continuous detection of dual-parameter.The different variables affecting the sensor performance were evaluated and optimized.Under the optimal conditions,the response curves,linear detection ranges,detection limits and response times of the dual-parameter sensor for testing sulfide and DCP were obtained,respectively.The sensor displays high selectivity,good repeatability and stability,which have good potentials in analyzing sulfide and DCP concentration of practical water samples.展开更多
基金The Faculty of Information Science and Technology,Universiti Kebangsaan Malaysia,provided funding for this research through the Research Grant“An Intelligent 4IR Mobile Technology for Express Bus Safety System Scheme DCP-2017-020/2”.
文摘One of the major causes of road accidents is sleepy drivers.Such accidents typically result in fatalities and financial losses and disadvantage other road users.Numerous studies have been conducted to identify the driver’s sleepiness and integrate it into a warning system.Most studies have examined how the mouth and eyelids move.However,this limits the system’s ability to identify drowsiness traits.Therefore,this study designed an Accident Detection Framework(RPK)that could be used to reduce road accidents due to sleepiness and detect the location of accidents.The drowsiness detectionmodel used three facial parameters:Yawning,closed eyes(blinking),and an upright head position.This model used a Convolutional Neural Network(CNN)consisting of two phases.The initial phase involves video processing and facial landmark coordinate detection.The second phase involves developing the extraction of frame-based features using normalization methods.All these phases used OpenCV and TensorFlow.The dataset contained 5017 images with 874 open eyes images,850 closed eyes images,723 open-mouth images,725 closed-mouth images,761 sleepy-head images,and 1084 non-sleepy head images.The dataset of 5017 images was divided into the training set with 4505 images and the testing set with 512 images,with a ratio of 90:10.The results showed that the RPK design could detect sleepiness by using deep learning techniques with high accuracy on all three parameters;namely 98%for eye blinking,96%for mouth yawning,and 97%for head movement.Overall,the test results have provided an overview of how the developed RPK prototype can accurately identify drowsy drivers.These findings will have a significant impact on the improvement of road users’safety and mobility.
基金supported by the Center for Cyber-Physical Systems,Khalifa University,under Grant 8474000137-RC1-C2PS-T5.
文摘The software engineering field has long focused on creating high-quality software despite limited resources.Detecting defects before the testing stage of software development can enable quality assurance engineers to con-centrate on problematic modules rather than all the modules.This approach can enhance the quality of the final product while lowering development costs.Identifying defective modules early on can allow for early corrections and ensure the timely delivery of a high-quality product that satisfies customers and instills greater confidence in the development team.This process is known as software defect prediction,and it can improve end-product quality while reducing the cost of testing and maintenance.This study proposes a software defect prediction system that utilizes data fusion,feature selection,and ensemble machine learning fusion techniques.A novel filter-based metric selection technique is proposed in the framework to select the optimum features.A three-step nested approach is presented for predicting defective modules to achieve high accuracy.In the first step,three supervised machine learning techniques,including Decision Tree,Support Vector Machines,and Naïve Bayes,are used to detect faulty modules.The second step involves integrating the predictive accuracy of these classification techniques through three ensemble machine-learning methods:Bagging,Voting,and Stacking.Finally,in the third step,a fuzzy logic technique is employed to integrate the predictive accuracy of the ensemble machine learning techniques.The experiments are performed on a fused software defect dataset to ensure that the developed fused ensemble model can perform effectively on diverse datasets.Five NASA datasets are integrated to create the fused dataset:MW1,PC1,PC3,PC4,and CM1.According to the results,the proposed system exhibited superior performance to other advanced techniques for predicting software defects,achieving a remarkable accuracy rate of 92.08%.
文摘A nursing care planning system that automatically generated nursing summaries from information entered into the Psychiatric Outcome Management System (PSYCHOMS?, Tanioka et al.), was developed to enrich the content of nursing summaries at psychiatric hospitals, thereby reducing the workload of nurses. Preparing nursing summaries entails finding the required information in nursing records that span a long period of time and then concisely summarizing this information. This time consuming process depends on the clinical experience and writing ability of the nurse. The system described here automatically generates the text data needed for nursing summaries using an algorithm that synthesizes patient information recorded in electronic charts, the Nursing Care Plan information or the data entered for North American Nursing Diagnosis Association (NANDA) 13 domains with predetermined fixed phrases. Advantages of this system are that it enables nursing summaries to be generated automatically in real time, simplifies the process, and permits the standardization of useful nursing summaries that reflect the course of the nursing care provided and its evaluation. Use of this system to automatically generate nursing summaries will allow more nursing time to be devoted to patient care. The system is also useful because it enables nursing summaries that contain the required information to be generated regardless of who prepares them.
基金supported by a grant from the Strategic Information and Communication R&D Promotion Program(SCOPE)in Japan(No.122309008).
文摘In psychiatric hospitals, the ratios between patients versus physician and patients versus nurse are low as compared to those in general hospitals. Furthermore, usages of electronic medical records are also low so that nurse administrators are limited in their ability to compile, analyze, and generate patient care staffing information for their administrative use. Psychiatric nurse administrators anticipate the development of a nursing administration analysis system that could perform personnel data simulation, manage information on nursing staff, and manage ward/ practice operations. Responding to this situation, the authors developed a nursing administration analysis system utilizing formulae from the Psychiatric Outcome Management System, PSYCHOMS®to aid nurse administrators. Such formulae are awaiting patent approval. The purpose of this study was to examine the validity of the formulae and the Structured Query Language (SQL) statement, and its practical effectiveness of analyzing data. The study findings showed that two kinds of computation expressions—a classification and extraction were able to display required information desired by nurse administrators. Moreover, significant information critical to assigning staff was validated to ensure high quality of nursing care according to the function and characteristic of the hospital ward.
基金supported by the Scientific Research Project of Xiangsihu College of Guangxi Minzu University,Grant No.2024XJKY06the National Natural Science Foundation of China under Grant No.U21A20464.
文摘The combined heat and power economic dispatch(CHPED)problem is a highly intricate energy dispatch challenge that aims to minimize fuel costs while adhering to various constraints.This paper presents a hybrid differential evolution(DE)algorithm combined with an improved equilibrium optimizer(DE-IEO)specifically for the CHPED problem.The DE-IEO incorporates three enhancement strategies:a chaotic mechanism for initializing the population,an improved equilibrium pool strategy,and a quasi-opposite based learning mechanism.These strategies enhance the individual utilization capabilities of the equilibrium optimizer,while differential evolution boosts local exploitation and escape capabilities.The IEO enhances global search to enrich the solution space,and DE focuses on local exploitation for more accurate solutions.The effectiveness of DE-IEO is demonstrated through comparative analysis with other metaheuristic optimization algorithms,including PSO,DE,ABC,GWO,WOA,SCA,and equilibrium optimizer(EO).Additionally,improved algorithms such as the enhanced chaotic gray wolf optimization(ACGWO),improved particle swarm with adaptive strategy(MPSO),and enhanced SCA with elite and dynamic opposite learning(EDOLSCA)were tested on the CEC2017 benchmark suite and four CHPED systems with 24,84,96,and 192 units,respectively.The results indicate that the proposed DE-IEO algorithm achieves satisfactory solutions for both the CEC2017 test functions and real-world CHPED optimization problems,offering a viable approach to complex optimization challenges.
基金funded by Ministry of Higher Education(MoHE)Malaysia,under Transdisciplinary Research Grant Scheme(TRGS/1/2019/UKM/01/4/2).
文摘The Cross-domain Heuristic Search Challenge(CHeSC)is a competition focused on creating efficient search algorithms adaptable to diverse problem domains.Selection hyper-heuristics are a class of algorithms that dynamically choose heuristics during the search process.Numerous selection hyper-heuristics have different imple-mentation strategies.However,comparisons between them are lacking in the literature,and previous works have not highlighted the beneficial and detrimental implementation methods of different components.The question is how to effectively employ them to produce an efficient search heuristic.Furthermore,the algorithms that competed in the inaugural CHeSC have not been collectively reviewed.This work conducts a review analysis of the top twenty competitors from this competition to identify effective and ineffective strategies influencing algorithmic performance.A summary of the main characteristics and classification of the algorithms is presented.The analysis underlines efficient and inefficient methods in eight key components,including search points,search phases,heuristic selection,move acceptance,feedback,Tabu mechanism,restart mechanism,and low-level heuristic parameter control.This review analyzes the components referencing the competition’s final leaderboard and discusses future research directions for these components.The effective approaches,identified as having the highest quality index,are mixed search point,iterated search phases,relay hybridization selection,threshold acceptance,mixed learning,Tabu heuristics,stochastic restart,and dynamic parameters.Findings are also compared with recent trends in hyper-heuristics.This work enhances the understanding of selection hyper-heuristics,offering valuable insights for researchers and practitioners aiming to develop effective search algorithms for diverse problem domains.
基金funded by Universiti Kebangsaan Malaysia,Fundamental Research Grant Scheme having Grant number FRGS/1/2023/ICT07/UKM/02/1Universiti Kebangsaan Malaysia Geran Universiti Penyelidikan having Grant number GUP-2024-009.
文摘The forthcoming 6G wireless networks have great potential for establishing AI-based networks that can enhance end-to-end connection and manage massive data of real-time networks.Artificial Intelligence(AI)advancements have contributed to the development of several innovative technologies by providing sophisticated specific AI mathematical models such as machine learning models,deep learning models,and hybrid models.Furthermore,intelligent resource management allows for self-configuration and autonomous decision-making capabilities of AI methods,which in turn improves the performance of 6G networks.Hence,6G networks rely substantially on AI methods to manage resources.This paper comprehensively surveys the recent work of AI methods-based resource management for 6G networks.Firstly,the AI methods are categorized into Deep Learning(DL),Federated Learning(FL),Reinforcement Learning(RL),and Evolutionary Learning(EL).Then,we analyze the AI approaches according to optimization issues such as user association,channel allocation,power allocation,and mode selection.Thereafter,we provide appropriate solutions to the most significant problems with the existing approaches of AI-based resource management.Finally,various open issues and potential trends related to AI-based resource management applications are presented.In summary,this survey enables researchers to understand these advancements thoroughly and quickly identify remaining challenges that need further investigation.
基金Project (Nos. 2001AA620104 and 2003AA603140) supported by theHi-Tech Research and Development Program (863) of China
文摘The behavioral responses of a tilapia (Oreochromis niloticus) school to low (0.13 mg/L), moderate (0.79 mg/L) and high (2.65 mg/L) levels of unionized ammonia (UIA) concentration were monitored using a computer vision system. The swimming activity and geometrical parameters such as location of the gravity center and distribution of the fish school were calculated continuously. These behavioral parameters of tilapia school responded sensitively to moderate and high UIA concen-tration. Under high UIA concentration the fish activity showed a significant increase (P<0.05), exhibiting an avoidance reaction to high ammonia condition, and then decreased gradually. Under moderate and high UIA concentration the school’s vertical location had significantly large fluctuation (P<0.05) with the school moving up to the water surface then down to the bottom of the aquarium alternately and tending to crowd together. After several hours’ exposure to high UIA level, the school finally stayed at the aquarium bottom. These observations indicate that alterations in fish behavior under acute stress can provide important in-formation useful in predicting the stress.
文摘Three kind of application of ADCP is reported for long-term monitoring in coastal sea.(1)The routine monitoring of water qualities. The water quality and ADCP echo data (600 kHz) observed in the long-term are analgzed at MT (Marine Tower) Station of Kansai International Airport in the Osaka Bay, Japan. The correlation between the turbidity and echo intensity in the surface layer is not good because air bubbles generated by breaking wave are not detected by the turbidity meter, but detected well by ADCP. When estimating the turbidity consists of plankton population from echo intensity, the effect of bubbles have to be eliminated. (2) Monitoring stirring up of bottom sediment. The special observation was carried out by using following two ADCP in the Osaka Bay, One ADCP was installed upward on the sea. The other ADCP was hanged downward at the gate type stand about 3 m above from the bottom. At the spring tide, high echo intensities indicating the stirring up of bottom sediment were observed. (3) The monitoring for the boundary condition of water mixing at an estuary. In summer season, the ADCP was set at the mouth of Tanabe Bay in Wakayama Prefecture, Japan. During the observation, water temperature near the bottom showed remarkable falls with interval of about 5-7 d. When the bottom temperature fell, the inflow current with low echo intensity water appears at the bottom layer in the ADCP record. It is concluded that when occasional weak northeast wind makes weak coastal upwelling at the mouth of the bay, the combination of upwelling with internal tidal flow causes remarkable water exchange and dispels the red tide.
基金a HKU Seed grant the Research Grants Council of the Hong Kong SAR(HKU7016/07P)
文摘In this article, the authors establish some new nonlinear difference inequalities in two independent variables, which generalize some existing results and can be used as handy tools in the study of qualitative as well as quantitative properties of solutions of certain classes of difference equations.
基金supported by National Natural Science Foundation of China (No. 60873208)
文摘Writable XOR executable （W⊕X） and address space layout randomisation （ASLR） have elevated the understanding necessary to perpetrate buffer overflow exploits [1] . However, they have not proved to be a panacea [1 3] , and so other mechanisms, such as stack guards and prelinking, have been introduced. In this paper, we show that host-based protection still does not offer a complete solution. To demonstrate the protection inadequacies, we perform an over the network brute force return-to-libc attack against a preforking concurrent server to gain remote access to a shell. The attack defeats host protection including W⊕X and ASLR. We then demonstrate that deploying a network intrusion detection systems （NIDS） with appropriate signatures can detect this attack efficiently.
文摘This paper presents a handwritten document recognition system based on the convolutional neural network technique.In today’s world,handwritten document recognition is rapidly attaining the attention of researchers due to its promising behavior as assisting technology for visually impaired users.This technology is also helpful for the automatic data entry system.In the proposed systemprepared a dataset of English language handwritten character images.The proposed system has been trained for the large set of sample data and tested on the sample images of user-defined handwritten documents.In this research,multiple experiments get very worthy recognition results.The proposed systemwill first performimage pre-processing stages to prepare data for training using a convolutional neural network.After this processing,the input document is segmented using line,word and character segmentation.The proposed system get the accuracy during the character segmentation up to 86%.Then these segmented characters are sent to a convolutional neural network for their recognition.The recognition and segmentation technique proposed in this paper is providing the most acceptable accurate results on a given dataset.The proposed work approaches to the accuracy of the result during convolutional neural network training up to 93%,and for validation that accuracy slightly decreases with 90.42%.
文摘One of the most difficult jobs in the post-genomic age is identifying a genetic disease from a massive amount of genetic data.Furthermore,the complicated genetic disease has a very diverse genotype,making it challenging to find genetic markers.This is a challenging process since it must be completed effectively and efficiently.This research article focuses largely on which patients are more likely to have a genetic disorder based on numerous medical parameters.Using the patient’s medical history,we used a genetic disease prediction algorithm that predicts if the patient is likely to be diagnosed with a genetic disorder.To predict and categorize the patient with a genetic disease,we utilize several deep and machine learning techniques such as Artificial neural network(ANN),K-nearest neighbors(KNN),and Support vector machine(SVM).To enhance the accuracy of predicting the genetic disease in any patient,a highly efficient approach was utilized to control how the model can be used.To predict genetic disease,deep and machine learning approaches are performed.The most productive tool model provides more precise efficiency.The simulation results demonstrate that by using the proposed model with the ANN,we achieve the highest model performance of 85.7%,84.9%,84.3%accuracy of training,testing and validation respectively.This approach will undoubtedly transform genetic disorder prediction and give a real competitive strategy to save patients’lives.
基金This work was supported by National Natural Science Foundation of China under Grant U21A20464,62066005Project of the Guangxi Science and Technology under Grant No.ZL23014016.
文摘This paper presents a Butterfly Optimization Algorithm(BOA)with a wind-driven mechanism for avoiding natural enemies known as WDBOA.To further balance the basic BOA algorithm's exploration and exploitation capabilities,the butterfly actions were divided into downwind and upwind states.The algorithm of exploration ability was improved with the wind,while the algorithm of exploitation ability was improved against the wind.Also,a mechanism of avoiding natural enemies based on Lévy flight was introduced for the purpose of enhancing its global searching ability.Aiming at improving the explorative performance at the initial stages and later stages,the fragrance generation method was modified.To evaluate the effectiveness of the suggested algorithm,a comparative study was done with six classical metaheuristic algorithms and three BOA variant optimization techniques on 18 benchmark functions.Further,the performance of the suggested technique in addressing some complicated problems in various dimensions was evaluated using CEC 2017 and CEC 2020.Finally,the WDBOA algorithm is used proportional-integral-derivative(PID)controller parameter optimization.Experimental results demonstrate that the WDBOA based PID controller has better control performance in comparison with other PID controllers tuned by the Genetic Algorithm(GA),Flower Pollination Algorithm(FPA),Cuckoo Search(CS)and BOA.
基金The Universiti Kebangsaan Malaysia(UKM)Research Grant Scheme FRGS/1/2020/ICT03/UKM/02/6 and GGPM-2020-028 funded this research.
文摘Many organizations apply cloud computing to store and effectively process data for various applications.The user uploads the data in the cloud has less security due to the unreliable verification process of data integrity.In this research,an enhanced Merkle hash tree method of effective authentication model is proposed in the multi-owner cloud to increase the security of the cloud data.Merkle Hash tree applies the leaf nodes with a hash tag and the non-leaf node contains the table of hash information of child to encrypt the large data.Merkle Hash tree provides the efficient mapping of data and easily identifies the changesmade in the data due to proper structure.The developed model supports privacy-preserving public auditing to provide a secure cloud storage system.The data owners upload the data in the cloud and edit the data using the private key.An enhanced Merkle hash tree method stores the data in the cloud server and splits it into batches.The data files requested by the data owner are audit by a third-party auditor and the multiowner authentication method is applied during the modification process to authenticate the user.The result shows that the proposed method reduces the encryption and decryption time for cloud data storage by 2–167 ms when compared to the existing Advanced Encryption Standard and Blowfish.
基金The Universiti Kebangsaan Malaysia(UKM)Research Grant Scheme GGPM-2020-028 funded this research.
文摘The fifth generation(5G)system is the forthcoming generation of the mobile communication system.It has numerous additional features and offers an extensively high data rate,more capacity,and low latency.However,these features and applications have many problems and issues in terms of security,which has become a great challenge in the telecommunication industry.This paper aimed to propose a solution to preserve the user identity privacy in the 5G system that can identify permanent identity by using Variable Mobile Subscriber Identity,which randomly changes and does not use the permanent identity between the user equipment and home network.Through this mechanism,the user identity privacy would be secured and hidden.Moreover,it improves the synchronization between mobile users and home networks.Additionally,its compliance with the Authentication and Key Agreement(AKA)structure was adopted in the previous generations.It can be deployed efficiently in the preceding generations because the current architecture imposes minimal modifications on the network parties without changes in the authentication vector’s message size.Moreover,the addition of any hardware to the AKA carries minor adjustments on the network parties.In this paper,the ProVerif is used to verify the proposed scheme.
文摘Autism spectrum disorder(ASD)is a challenging and complex neurodevelopment syndrome that affects the child’s language,speech,social skills,communication skills,and logical thinking ability.The early detection of ASD is essential for delivering effective,timely interventions.Various facial features such as a lack of eye contact,showing uncommon hand or body movements,bab-bling or talking in an unusual tone,and not using common gestures could be used to detect and classify ASD at an early stage.Our study aimed to develop a deep transfer learning model to facilitate the early detection of ASD based on facial fea-tures.A dataset of facial images of autistic and non-autistic children was collected from the Kaggle data repository and was used to develop the transfer learning AlexNet(ASDDTLA)model.Our model achieved a detection accuracy of 87.7%and performed better than other established ASD detection models.Therefore,this model could facilitate the early detection of ASD in clinical practice.
文摘Escalating cyber security threats and the increased use of Internet of Things(IoT)devices require utilisation of the latest technologies available to supply adequate protection.The aim of Intrusion Detection Systems(IDS)is to prevent malicious attacks that corrupt operations and interrupt data flow,which might have significant impact on critical industries and infrastructure.This research examines existing IDS,based on Artificial Intelligence(AI)for IoT devices,methods,and techniques.The contribution of this study consists of identification of the most effective IDS systems in terms of accuracy,precision,recall and F1-score;this research also considers training time.Results demonstrate that Graph Neural Networks(GNN)have several benefits over other traditional AI frameworks through their ability to achieve in excess of 99%accuracy in a relatively short training time,while also capable of learning from network traffic the inherent characteristics of different cyber-attacks.These findings identify the GNN(a Deep Learning AI method)as the most efficient IDS system.The novelty of this research lies also in the linking between high yielding AI-based IDS algorithms and the AI-based learning approach for data privacy protection.This research recommends Federated Learning(FL)as the AI training model,which increases data privacy protection and reduces network data flow,resulting in a more secure and efficient IDS solution.
基金supported by Dana Impak Perdana fund,no.UKM DIP-2018-040 and Fundamental Research Grant Scheme fund no FRGS/1/2018/TK04/UKM/02/7 under Author R.Hassan.
文摘In networking,one major difficulty that nodes suffer from is the need for their addresses to be generated and verified without relying on a third party or public authorized servers.To resolve this issue,the use of selfcertifying addresses have become a highly popular and standardized method,of which Cryptographically Generated Addresses(CGA)is a prime example.CGA was primarily designed to deter the theft of IPv6 addresses by binding the generated address to a public key to prove address ownership.Even though the CGA technique is highly effective,this method is still subject to several vulnerabilities with respect to security,in addition to certain limitations in its performance.In this study,the authors present an intensive systematic review of the literature to explore the technical specifications of CGA,its challenges,and existing proposals to enhance the protocol.Given that CGA generation is a time-consuming process,this limitation has hampered the application of CGA in mobile environments where nodes have limited energy and storage.Fulfilling Hash2 conditions in CGA is the heaviest and most timeconsuming part of SEND.To improve the performance of CGA,we replaced the Secure Hash Algorithm(SHA1)with the Message Digest(MD5)hash function.Furthermore,this study also analyzes the possible methods through which a CGA could be attacked.In conducting this analysis,Denial-of-Service(DoS)attacks were identified as the main method of attack toward the CGA verification process,which compromise and threaten the privacy of CGA.Therefore,we propose some modifications to the CGA standard verification algorithm to mitigate DoS attacks and to make CGA more security conscious.
基金Funded by the Natural Science Foundation of Hubei Province(No.2022CFB861)the Wenhua College Research and Innovation Team(No.2022T01)。
文摘A novel fiber optic sensor based on hydrogel-immobilized enzyme complex was developed for the simultaneous measurement of dual-parameter,the leap from a single parameter detecting fiber optic sensor to a fiber optic sensor that can continuously detect two kinds of parameters was achieved.By controlling the temperature from high to low,the function of fiber sulfide sensor and fiber DCP sensor can be realized,so as to realize the continuous detection of dual-parameter.The different variables affecting the sensor performance were evaluated and optimized.Under the optimal conditions,the response curves,linear detection ranges,detection limits and response times of the dual-parameter sensor for testing sulfide and DCP were obtained,respectively.The sensor displays high selectivity,good repeatability and stability,which have good potentials in analyzing sulfide and DCP concentration of practical water samples.
