Search

The purpose of designing embedded Internet of things and studying the ad hoc technology of gateway is to meet the monitoring requirements brought by the continuous expansion of data center. Among them, in the aspect of dynamic data monitoring, this paper designs a dynamic acquisition algorithm, that is, by analyzing the characteristics of data center resources and the number of information acquisition methods, so as to adjust the data acquisition cycle and to reduce the monitoring workload and improve the monitoring flexibility. At the same time, this paper also studies the automatic psychological evaluation system of students. It is found that with the intensification and efficient operation of social competition, some special groups of college students are facing increasing pressure, and events caused by psychological problems emerge one after another. Even though there are many psychological evaluation systems on the network, a large number of psychological knowledge and mental health management on the Internet cannot effectively manage and improve the psychological status of college students. Therefore, the university needs to build a psychological evaluation system for college students in time and combine the automatic psychological evaluation system for college students with embedded Internet of things and cloud service technology to improve the accuracy of psychological evaluation. The application of the system, college students can independently evaluate their own mental state, if there are psychological problems can be timely counseling, not only can improve the level of mental health education in colleges and universities, but also can promote the development of physical and mental health. [ABSTRACT FROM AUTHOR]

With the continuous progress of Internet technology, so there is a higher demand for English translation. In this context, this paper realizes the design and improvement of the online English translation system by combining the image segmentation algorithm. After the experimental test of the algorithm of this system, if there is a lot of salt-and-pepper noise in the target image in the actual system application process, the method designed in this paper can reflect a better processing effect. Based on the analysis of performance parameters, this method can effectively process Gaussian noise images with high precision. After the completion of the system design, this paper tests the system. From the test and operation results, the software has reached the design objectives and completed the work tasks. The English translation online teaching system can form a safe, stable and high-bandwidth English translation teaching network and can realize the connection of the whole network. Users can share various software and hardware resources, completed the design of a translation platform that can realize rapid English communication, share experience and discuss, to effectively improve the learner's initiative. In this paper, an effective online teaching system is designed by introducing image segmentation algorithm into the field of English translation. [ABSTRACT FROM AUTHOR]

Data mining algorithms can process target data and extract useful hidden information, which is helpful for decision making. However, current mining algorithms have some shortcomings such as time-consuming processing of big data or inability to process massive data. Since data mining technology cannot be used in the traditional cloud platform environment, it is necessary to improve the algorithm to make it more suitable for the cloud platform environment. By analyzing the actual application process of BP classification algorithm, this paper expounds the practicability of BP classification algorithm, analyzes the data mining process based on Hadoop cloud platform, and expounds the development ideas of BP classification algorithm. The source of data mining algorithm supported by cloud computing is discussed. Finally, Hadoop cloud platform. This paper designed the corresponding system architecture and data interface, and established a suitable test environment for this system, and completed the simulation experiment test through the design. The calculation time of this algorithm is proportional to the amount of data, showing a linear relationship. In data mining, the optimized BP algorithm in this paper can significantly save resources in terms of spatial features. This paper designs an optimized operating system based on Hadoop platform through comprehensive analysis and improvement algorithm. [ABSTRACT FROM AUTHOR]

When constructing the algorithm model of sports training action classification, the accuracy of action classification has an important impact on the algorithm model. How to improve the algorithm model to improve the accuracy of sports training action classification needs further research. Based on BP neural network algorithm, this paper carries out the modeling of BP neural network signal classification algorithm and the construction of BP neural network, and deduces the BP algorithm in detail. Firstly, this paper applies genetic algorithm to the initial parameter selection of BP algorithm to avoid the local optimization problem. When carrying out chromosome coding, binary coding is easy to cause the problem of too long coding string, which also needs to be restored and decoded. The algorithm runs too long and the learning accuracy is not high. Therefore, this paper uses real coding. Through simulation analysis, it can be seen that the classification accuracy of the improved algorithm model is significantly higher than that of the simple BP algorithm. In addition, this paper analyzes the requirements of the sports training management visualization system, introduces the system structure framework and network topology, describes in detail the functions of the user information management module, the training plan management module, the training test management module, the competition information management module and the scientific research information management module, and tests the visualization function of the system. Finally, this paper analyzes the problems existing in the current sports training management, and puts forward the development strategy of sports training management based on this, which lays a theoretical foundation for the scientific development of sports training. [ABSTRACT FROM AUTHOR]

In recent years, the cuckoo search (CS) algorithm has been successfully applied to single-objective optimization problems. But in real life, most optimization problems are multi-objective optimization problems (MOPs). In order to enable CS to better solve MOPs, this paper proposes an elite-guided multi-objective cuckoo search algorithm based on cross-operation and information enhancement (CIE-MOCS). This algorithm first enhances its population diversity through crossover operation, then adds elite individuals to guide its update process to speed up the algorithm convergence speed. Finally, the method of information enhancement is adopted in the abandonment process, so that the algorithm is not easy to fall into the local optimum. In order to verify the performance of the algorithm, this paper uses a variety of benchmark functions and performance evaluation indicators to evaluate it, and provides a case to verify the effectiveness of the algorithm in practical applications. The experimental results show that CIE-MOCS has good performance compared with the contrasting algorithms. [ABSTRACT FROM AUTHOR]

In the lung computer-aided detection (Lung CAD) system, the region of interest (ROI) of lung nodules has more false positives, making the imbalance between positive and negative (true positive and false positive) samples more likely to lead to misclassification of true positive nodules, a cost-sensitive multikernel learning support vector machine (CS-MKL-SVM) algorithm is proposed. Different penalty coefficients are assigned to positive and negative samples, so that the model can better learn the features of true positive nodules and improve the classification effect. To further improve the detection rate of pulmonary nodules and overall recognition accuracy, a score function named F-new based on the harmonic mean of accuracy (ACC) and sensitivity (SEN) is proposed as a fitness function for subsequent particle swarm optimization (PSO) parameter optimization, and a feasibility analysis of this function is performed. Compared with the fitness function that considers only accuracy or sensitivity, both the detection rate and the recognition accuracy of pulmonary nodules can be improved by this new algorithm. Compared with the grid search algorithm, using PSO for parameter search can reduce the model training time by nearly 20 times and achieve rapid parameter optimization. The maximum F-new obtained on the test set is 0.9357 for the proposed algorithm. When the maximum value of F-new is achieved, the corresponding recognition ACC is 91%, and SEN is 96.3%. Compared with the radial basis function in the single kernel, the F-new of the algorithm in this paper is 2.16% higher, ACC is 1.00% higher and SEN is equal. Compared with the polynomial kernel function in the single kernel, the F-new of the algorithm is 3.64% higher, ACC is 1.00% higher and SEN is 7.41% higher. The experimental results show that the F-new, ACC and SEN of the proposed algorithm is the best among them, and the results obtained by using multikernel function combined with F-new index are better than the single kernel function. Compared with the MKL-SVM algorithm of grid search, the ACC of the algorithm in this paper is reduced by 1%, and the results are equal to those of the MKL-SVM algorithm based on PSO only. Compared with the above two algorithms, SEN is increased by 3.71% and 7.41%, respectively. Therefore, it can be seen that the cost sensitive method can effectively reduce the missed detection of nodules, and the availability of the new algorithm can be further verified. [ABSTRACT FROM AUTHOR]

With the gradual advancement of image sensor technology and the gradual complexity of the application environment, single-source imaging sensors have been difficult to meet practical application requirements. Therefore, the registration and fusion techniques of infrared images and visible images are hot topics in recent years. This article aims to study the fusion of two images. Based on the traditional image fusion algorithm, the color space IHS transform and the lifting wavelet transform are combined according to the characteristics of infrared image and visible light image captured by the UAV imaging device. The experimental results show that the algorithm can not only retain the brightness information of the infrared target. Among them, some details of the visible light image are retained. In this paper, the four objective criteria of information entropy, standard deviation, image mean and average gradient are used to evaluate the fusion effect of the proposed algorithm. The obtained values are 7.285, 29.487, 122.2739, and 4.5678, respectively, which are higher than other algorithms. It can be seen that the algorithm proposed in this paper has important practical significance. [ABSTRACT FROM AUTHOR]

Focusing on non-iterative approaches in training feed-forward neural networks, this special issue includes 12 papers to share the latest progress, current challenges, and potential applications of this topic. This editorial presents a background of the special issue and a brief introduction to the 12 contributions. [ABSTRACT FROM AUTHOR]

Recent developments in fuzzy logic have presented new theories, concepts and algorithms extending the original ideas of the pioneering work of Prof. Zadeh. In this case, extensions to type-1 like interval type-2 fuzzy logic, general type-2 fuzzy logic, intuitionistic fuzzy logic and other similar approaches have been recently proposed to tackle more complex real-world problems. In the 11th paper, by Fevrier Valdez and Cinthia Ochoa, a new approach for dynamic parameter adaptation in the harmony search algorithm for the optimization of interval type-2 fuzzy logic controllers was outlined with interesting results in the optimization of benchmark problems. [Extracted from the article]

Two goals of multi-objective evolutionary algorithms are effectively improving their convergence and diversity and making the Pareto set evenly distributed and close to the real Pareto front. At present, the challenges to be solved by the multi-objective evolutionary algorithm are to improve the convergence and diversity of the algorithm, and how to better solve functions with complex PF and/or PS shapes. Therefore, this paper proposes a gray wolf optimization-based self-organizing fuzzy multi-objective evolutionary algorithm. Gray wolf optimization algorithm is used to optimize the initial weights of the self-organizing map network. New neighborhood relationships for individuals are built by self-organizing map, which can maintain the invariance of feature distribution and map the structural information of the current population into Pareto sets. Based on this neighborhood relationship, this paper uses the fuzzy differential evolution operator, which constructs a fuzzy inference system to dynamically adjust the weighting parameter in the differential operator, to generate a new initial solution, and the polynomial mutation operator to refine them. Boundary processing is then conducted. Experiments on 15 problems of GLT1-6 and WFG1-9 and the algorithm proposed in this paper achieve the best on 18 values. And the result shows that the convergence and diversity of the proposed algorithm are better than several state-of-the-art multi-objective evolutionary algorithms. [ABSTRACT FROM AUTHOR]

In order to solve the problems of low learning efficiency and slow convergence speed when mobile robot uses reinforcement learning method for path planning in complex environment, a reinforcement learning method based on each round path planning result is proposed. Firstly, the algorithm adds obstacle learning matrix to improve the success rate of path planning; and introduces heuristic reward to speed up the learning process by reducing the search space; then proposes a method of dynamically adjusting the exploration factor to balance the exploration and utilization in path planning, so as to further improve the performance of the algorithm. Finally, the simulation experiment in grid environment shows that compared with Q-learning algorithm, the improved algorithm not only shortens the average path length of the robot to reach the target position, but also speeds up the learning efficiency of the algorithm, so that the robot can find the optimal path more quickly. The code of EPRQL algorithm proposed in this paper has been published to GitHub: https://github.com/panpanpanguoguoqian/mypaper1.git. [ABSTRACT FROM AUTHOR]

The emergence of computer interactive teaching system greatly facilitates the needs of students to learn anytime and anywhere at this stage. At the same time, the increase in online course video, voice, files and other teaching resources also puts forward higher requirements for the operation of the teaching system. This article addresses the practical needs of online teaching and proposes a scalable computer interactive teaching system designed with the aid of large-scale multimedia data analysis algorithms. The system utilizes a principal component analysis linear data feature extraction algorithm to extract data features and implements data prediction through linear transformation methods for data reconstruction. This approach enables the system to effectively extract and analyze data, while also facilitating efficient data prediction for improved teaching outcomes. With its scalability and advanced data analysis capabilities, this system represents a promising solution for enhancing the effectiveness of online teaching and improving the overall learning experience for students. Most online course resources are displayed in the form of multimedia. Therefore, this paper uses robust component analysis to process multimedia data, especially for some problems such as data missing and multimedia data interference. This paper uses RC ranking-related evaluation method and R-value evaluation method to evaluate the prediction data, which proves that parameter selection plays an important role in controlling sparse regression. The algorithm proposed in this paper was tested using actual video and picture data related to the teaching system resources. The results indicated that the algorithm achieved the best data recognition performance when the number of theme models was set to 100. This finding provides an important reference for the system design. Finally, this paper designs a computer interactive teaching system which includes three modules: administrator, student and teacher, and tests the system, which proves the practical operation possibility and application value of the system. [ABSTRACT FROM AUTHOR]

Due to the large number of stock companies, complex stock categories and inconsistent evaluation standards of market value, it is not conducive to the choice of investors. Based on the theory and practice of traditional stock performance evaluation model, this paper integrates the algorithm thought structure of Fama–French five-factor model and proposes a machine learning algorithm model for stock performance research. In addition, this paper also builds a model that can evaluate the style and timing ability of fund managers to improve the fund performance evaluation system to a greater extent. With the help of the performance evaluation updating function of the model, it provides a new experience material in the empirical research composition of the fund performance field. In the system test module, the former prediction data and the actual experimental data are integrated, and the two are sorted out and compared, which proves the feasibility and effectiveness of the proposed algorithm model. The final experimental results verify the usefulness of the model in stock return prediction and analysis of influencing factors. [ABSTRACT FROM AUTHOR]

Based on the rapid development of big data, cloud computing, Internet of things and other technologies in recent years, intelligent hardware devices has been applied to all aspects of life. Under this background, some scholars have put forward relevant concepts such as "Smart Life". In the field of mass sports life, through the development and application of science and technology, there has been new changes related to the application of neural network algorithm technology and intelligent hardware devices. Therefore, artificial intelligence wearable devices based on wearable technology came into being. This paper analyzes the application of this device in mass sports activities. Then, this paper describes the key research technologies of motion data processing based on neural network algorithm, including: depth frame differential convolution neural network structure, motion data extraction method, human motion signal processing algorithm, etc.; then it analyzes the action recognition and interaction system design based on Intelligent wearable devices. Finally, it analyzes the recognition results of human action system, the accuracy of human action recognition system and the factors that affect the performance of the recognition system. It is concluded that the artificial intelligent wearable devices designed in this paper can be well used in popular sports activities. Finally, it introduces the research on the evaluation strategy of popular sports activities based on artificial intelligence, and hopes that this equipment can help public sports activities. This paper studies the neural network algorithm and applies it to the design process of artificial intelligence wearable devices, which promotes the development of mass sports activity evaluation. [ABSTRACT FROM AUTHOR]

To expand the application scope of the WCA algorithm in the actual process, this research has optimized the problems in the algorithm in two aspects: The first aspect further improves the operating mechanism of the WCA algorithm and improves the performance of optimization problem. In the second aspect, other optimization strategy mechanisms of the WCA algorithm are introduced to enable the algorithm to optimize multi-objective and multidimensional problems. This paper studies a physical test system and sports intelligent based on hybrid wolf pack algorithm and IoT. The system collects and sends data from the data collection terminal of the Web server, receives the data through the wireless module, and sends it to the Web server through the network. The Web server processes and stores the data information to generate a database, and users can view their own sports information by logging in to the Web service program with the account password. In addition, the teacher and administrator accounts have the ability to view all users' exercise information. The system adds three sports items, pull-ups, squats, and standing long jumps. At the same time, the system uses a general motion recognition algorithm, which can effectively reuse and add new sports items. According to the actual needs of intelligent sports training, this paper combines somatosensory technology, bone tracking technology, and motion recognition algorithm to realize a high-precision, low-latency intelligent sports training system. [ABSTRACT FROM AUTHOR]

Nowadays, physical dance is widely spread in the society as an emerging sport. Dance movement is favored by people because of its unique social function and fitness effect. For dance teaching, dance movement analysis can help optimize and improve the existing dance movements and the understanding and inheritance of traditional dance movements. With the rise of online teaching, intelligent identification and analysis of dance movements can promote the better development of sports dance teaching. However, the relevant research in this area is still very scarce. As the basis of this kind of research, there is an urgent need for dance motion recognition technology. Based on this background, this paper by introducing neural network algorithm for dance teaching sports image special recognition design, the algorithm can combine feature extraction technology to process video, extract the dance movements in the target data set, then for the extraction of cumulative feature extraction operation, in order to accumulate all the collected target features, so as to further complete the gradient histogram acquisition. Through the design experimental test, the cumulative feature image extraction results obtained through the algorithm are obviously better than the traditional image recognition results, so the design rationality and effectiveness of the algorithm are proved, and the sports dance teaching can be specially assisted. This paper designs an effective auxiliary image recognition algorithm by introducing the neural network algorithm into the field of sports dance teaching. [ABSTRACT FROM AUTHOR]

A profound concern dwelling on developing practical control algorithms for autonomous cars is to guarantee robustness and resilience in harsh driving situations. Extraneous environmental factors coupled with structured and unstructured uncertainties have always posed concerns about the proposed algorithms' effectiveness. This paper presents an improved control algorithm based on a robust adaptive neural network trained with integral sliding mode (NN-ISMC) with the capacity for state estimation and auxiliary control inputs. Additionally, the paper considers the contribution of active front steering (AFS) integrated with direct yaw moment control (DYC) to employ during critical driving scenarios. Moreover, a super-twisting ESO disturbance observer (STESO-DO) was developed to estimate disturbances imposed on the car during an emergency maneuver, such as double-lane change, in terms of successive strong gusts of crosswind. Additional uncertainties were introduced to the system to evaluate the algorithm's robustness, such as the tire cornering stiffness and traveling speed. The performance of the designed framework was further assessed against two documented comparable methods in the literature using high-fidelity MATLAB/Simulink–CarSim co-simulations. The findings from various driving conditions and speeds indicate that the proposed controller successfully stabilizes the handling dynamics and thus enhances the path-tracking performance compared to the previously reported methods. [ABSTRACT FROM AUTHOR]

Label-specific features learning can effectively exploit the unique features of each label, which alleviates the high dimensionality and improves the classification performance of multi-label. However, most existing label-specific features learning algorithms assume that label space is complete, ignoring the effect of missing labels on the classification accuracy. Some methods try to recover the missing labels first and then learn the mapping between the completed label matrix and the feature matrix. However, early intervention in the recovery of missing labels may affect the distribution of original labels to a certain extent. In this paper, feature-label dual-mapping for missing label-specific features learning is proposed. According to the information that the label depends on the feature, the dual-mapping weight of the complete feature space and the missing label space is jointly learned. Therefore, the proposed algorithm is to conduct latent missing labels recovery by feature-label dual-mapping to directly obtain target weight in this paper, avoiding the negative influence of early label recovery intervention. Compared with several state-of-the-art methods in 10 benchmark multi-label data sets, the results show that the proposed algorithm is reasonable and effective. [ABSTRACT FROM AUTHOR]

At present, the development of value chain mainly shows that the added value is positive; otherwise, the industrial value chain will be eliminated. Therefore, the construction and measurement of the industrial value chain height can effectively avoid the collapse of the industrial value chain caused by the low system value of the value chain. In this paper, the mobile virtualization channel allocation algorithm is used to construct the multi-objective industrial value chain height index, and the actual industrial value chain height in China is measured. This paper first describes the channel allocation algorithm of mobile virtualization, extracts the allocation rules of the algorithm, constructs the model of multi-objective industrial value chain, then measures the value chain of eight regions in China combined with the elements of technology, production, and marketing in the height of value chain, and quantitatively compares the results of value chain height of different regions according to the embeddedness and depth. Finally, it puts forward the improvement scheme for the measurement results and the problems faced by the value chain. The research shows that the average value chain height of South China is about 70.86, while the average value chain height of Northwest China is about 46.58, ranging from 40.58 to 60.45. Through quantitative data, the value chain height of South China is generally higher than that of Northwest China by more than 50%, which may be related to the innovation of technology and the imitation form of production. Therefore, this paper proposes to strengthen the technological innovation in the upstream of the value chain, so as to drive the development of the regional value chain. [ABSTRACT FROM AUTHOR]

In this paper, Protist algorithm (PA) and Otocyon megalotis optimization algorithm (OOA) are applied to solve the power loss lessening problem. Protist Algorithm (PA) is modelled based on the Protist's natural activities. Protist exists in moist places. The leading nutritious phase is Plasmodium, the energetic and vibrant phase of Protist. In this segment, the organic substance in Protist search for food in surroundings and conceals enzymes for digestion. Natural actions of Otocyon megalotis are emulated to design the OOA approach. In the projected OOA searching of regions in exploration, for foodstuff the Otocyon megalotis mark the prey in the space is indicated as a global exploration. Real power loss reduction and Voltage stability enhancement are the key objectives of the paper. To solve the problem, Protist algorithm (PA) and Otocyon megalotis optimization algorithm. In the course of the migration procedure, the anterior end outspreads and interconnected arterial system that authorize cytoplasm to stream inside. Then, mutation and cross-over probability are employed to augment the performance of the Protist algorithm (PA). With this integration engendering of the population is done. Mutation classes the population exploration agents (PN) in uphill order conferring to the agents appropriateness (fitness) cost. Consequently, the technique splits the organized agents into three fragments rendering to their fitness value. In which PN/3 denotes to the population possess pre-eminent (aptness) fitness values, subsequently with second pre-eminent and poorest aptness (fitness) values. Then, in this paper, Otocyon megalotis optimization algorithm (OOA) is applied for solving the Power loss lessening problem. In the subsequent segment, navigate during the haunt to seal prey previous to the hit was replicated as a local search. In exploration, the data obtained is shared to all the associates of the family unit for continued existence and growth. Examination of the nearby terrain is modelled with reference to the fitness of all entities. Most excellent entity has investigated the majority fascinating terrain and it will be shared with family unit of Otocyon megalotis. Primarily, Otocyon megalotis show that it not involved in hunting. Conversely, as soon as moving near to prey Otocyon megalotis will perform the attack in quick mode. This approach imitated and designed in the local search segment. Authenticity of the Protist algorithm (PA) and Otocyon megalotis optimization algorithm (OOA) is corroborated in 23 benchmark functions, IEEE 30, 57, 300 and 354 test systems. Power Loss reduction achieved with voltage stability enhancement. Real power loss reduction attained. Both the Protist algorithm (PA) and Otocyon megalotis optimization algorithm (OOA) performed well in solving the Power loss reduction problem. [ABSTRACT FROM AUTHOR]

Aiming at the shortcomings of Chimp optimization algorithm (ChOA), which is easy to fall into local optimal value and imbalance between global exploration ability and local exploitation ability. To improve ChOA from the perspective of multi-strategy mixing, MSChimp was proposed, and the algorithm was applied to global optimization and minimum spanning tree problems. The main research work of this paper is as follows: (1) In the initialization stage of ChOA, an opposition-based learning strategy was introduced to improve the population diversity; Sine Cosine Algorithm (SCA) was introduced in the exploitation process to improve the convergence speed and accuracy of the algorithm in the later stage, so as to balance the exploration and exploitation capabilities of the algorithm. (2) The improved algorithm was compared with different types of meta-heuristic algorithms in 20 benchmark functions and CEC 2019 test sets, and was used to solve the minimum spanning tree. The experimental results show that the improved ChOA has significantly improved the ability to find the optimal value, which verifies the effectiveness and feasibility of MSChimp. Compared with other algorithms, the algorithm proposed in this paper has strong competitiveness. [ABSTRACT FROM AUTHOR]

This paper presents an improved arithmetic optimization algorithm that incorporates hybrid elite pool strategies to address the limitations of the arithmetic optimization algorithm (AOA). In AOA, the linear mathematical optimization acceleration (MOA) function cannot balance global exploitation and local exploration well. Therefore, the accuracy and convergence speed of the algorithm cannot be guaranteed. To improve the performance of AOA, this paper reconstructed a nonlinear MOA function, which is expected to balance the exploitation and the exploration of AOA. Furthermore, four hybrid elite pool strategies are integrated to enhance the ability to escape local optima. The proposed algorithm inherits the fast convergence of AOA and develops the performance of escaping local optima. Numerical experiment results on benchmark functions and engineering problems show that the proposed algorithm outperforms other compared meta-heuristic algorithms in terms of convergence speed and accuracy. [ABSTRACT FROM AUTHOR]

This paper investigates and analyzes the dynamics of the two-dimensional Duffing map. Multistability behavior has been observed from the system numerically. Such behavior, especially the coexistence of chaotic and periodic attractors, is undesirable in the applications of chaos-based cryptography. Therefore, we design and implement a Sine–Cosine chaotification technique to enhance chaos in the multi-stable regions. Furthermore, this paper proposes a new image encryption algorithm to examine the performance of the generalized Duffing map in cryptography applications. Simulation results and security analysis reveal that the proposed algorithm can effectively encrypt and decrypt several image types with a high level of security. [ABSTRACT FROM AUTHOR]

The continuous progress of society has created conditions for the widespread use of information technology. People rely more and more on information technology and the Internet. People can use the Internet to retrieve relevant information to meet their work needs, but the widespread use of the Internet is also accompanied by network security issues. The existence of network security problems will affect people's work to varying degrees. Network security management departments use intrusion detection technology and set firewalls to improve the security of people's personal information on the Internet and prevent viruses and some criminals from stealing people's important information through the Internet, causing adverse effects on the Internet environment. In the daily management of the school, the extensive use of the campus network can not only facilitate students' daily study and life, but also improve the efficiency of school management. Ensuring the network security of campus network is an important work to ensure the campus management and students' study and life. The four parts of network security work include setting firewall, encrypting cloud data, using intrusion detection technology, and recovering data. Intrusion detection technology is extremely important for the development of network security work, which can help people actively find vulnerabilities in networks and systems, identify possible behaviors that may invade systems and networks, and give early warning or automatically close intrusion channels. There are various types of cloud data, so cloud database plays an extremely important role in the development of various fields. This paper uses intrusion detection algorithms to design the campus network security protection system, which provides security for the use of the campus network. [ABSTRACT FROM AUTHOR]

Today, data storage technology is also gradually improving. Various industries can store massive amounts of data for analysis. The global climate change and the bad ecology led to frequent occurrence of natural disasters. Therefore, it is necessary to establish an effective emergency materials distribution system. The neural network model is used to calculate and the optimal emergency distribution route is analyzed according to the historical information and the data. Considering backpropagation, this paper further disposing a method to further improve the calculation of neural network algorithm. From the perspective of structural parameters of neural network algorithms, this paper uses genetic algorithms to construct predictions, and combines the actual purpose of material distribution after disasters. Considering the capacity constraints of distribution centers, time constraints, material needs of disaster relief points and different means of transportation, a dual-objective path planning with multiple distribution centers and multiple disaster relief points with the shortest overall delivery time and lowest overall delivery cost is constructed. By establishing an emergency material distribution system, it can maximize the prompt and accurate delivery after a natural disaster occurs, and solves the urgent needs of the people. [ABSTRACT FROM AUTHOR]

This paper proposes PatternAttack to mitigate two major issues of Adversarial Transformation Network (ATN) including the low diversity and the low quality of adversarial examples. In order to deal with the first issue, this research proposes a stacked convolutional autoencoder based on patterns to generalize ATN. This proposed autoencoder could support different patterns such as all-pixel pattern, object boundary pattern, and class model map pattern. In order to deal with the second issue, this paper presents an algorithm to improve the quality of adversarial examples in terms of L 0 -norm and L 2 -norm. This algorithm employs adversarial pixel ranking heuristics such as JSMA and COI to prioritize adversarial pixels. To demonstrate the advantages of the proposed method, comprehensive experiments have been conducted on the MNIST dataset and the CIFAR-10 dataset. For the first issue, the proposed autoencoder generates diverse adversarial examples. For the second issue, the proposed algorithm significantly improves the quality of adversarial examples. In terms of L 0 -norm, the proposed algorithm decreases from hundreds of adversarial pixels to one adversarial pixel. In terms of L 2 -norm, the proposed algorithm reduces the average distance considerably. These results show that the proposed method can generate high-quality and diverse adversarial examples in practice. [ABSTRACT FROM AUTHOR]

With the development of the big data era, the manufacturing industry is focusing on integrating the technologies related to the evolving intelligent industry. Intelligent manufacturing is the latest form of advanced manufacturing development. In order to pay attention to the design and evaluation status of manufacturing system, we understand the connotation of manufacturing system from the perspective of actual manufacturing production and sales. In this paper, the evaluation model and algorithm of Intelligent Manufacturing System Based on pattern recognition and big data are proposed. In the process of studying the information transmission of intelligent manufacturing system, the evaluation hierarchy algorithm of intelligent control system is improved with reference to several modeling methods. Finally, an intelligent manufacturing system suitable for processing big data is designed. In the experiment in workshop A, with the help of computer simulation of artificial intelligent activities, the intelligent manufacturing system analyzes the processing and production data, obtains the dynamic requirements of products in the machining workshop, and provides judgment basis for the construction of dynamic units in enterprises. The experimental results show that after the change of the production organization model using the seasonal change forecasting method, the introduction of the smart manufacturing system, the sales production of this shop has been increasing, and in the 20th month has increased seven times more than the initial, while the overall situation of the predicted sales and actual sales and costs, the superiority of using the smart manufacturing evaluation model is derived. [ABSTRACT FROM AUTHOR]

The emerging technologies such as high-performance computing, cloud computing and large data centre networks demand very high injection rate, high scalability and very low switching latency (SL) that can manage high network traffic, and buffering and also be cost-effective, so that they can be a viable solution for future manufacturing industry. To overcome the biggest challenge of optical buffering and big quantity of data switching, the Data Vortex (DV) all optical switch innovations and its various enhancements such as Augmented Data Vortex (ADV) seem to be the possible solution and can bring the revolution in the industrial world. In this paper, a new generalized algorithm has been used to simulate the equivalent planar virtual model of 3-D ADV and is named as equivalent planar algorithm (EPA). The algorithm is more versatile and provides more flexible design for multiple level networks. To verify the proposed EPA of ADV (or EPA ADV), the simulations have been performed for various input traffic loads and switch size keeping active angle 'A' constant. Also the ADV switch is not studied under asymmetric mode (AM) of operation till now which has been done here. Lastly the paper also provides the performance comparison study of original DV and EPA ADV under symmetric and asymmetric I/O mode. The results obtained from simulation have shown that under modest degree of asymmetric I/O operation of the switch EPA ADV shows performance improvement in injection rate (throughput) and average switch latency (SL) under different traffic pattern as compared to symmetric mode (SM) of operation in ADV. [ABSTRACT FROM AUTHOR]

It is hard to come up with a strong learning algorithm with high cross-media retrieval accuracy, but finding a weak learning algorithm with slightly higher accuracy than random prediction is simple. This paper proposes an innovative Bagging-based cross-media retrieval algorithm (called BCMR) based on this concept. First, we use bootstrap sampling to take a random sample from the original set. The amount of sample abstracted by bootstrapping is set to be the same as the amount of sample abstracted by the original dataset. Secondly, 50 bootstrap replicates are used to train 50 weak classifiers. In our experiments, we used homogeneous individual classifiers and eight different baseline methods. Last but not least, we generate the final strong classifier from 50 weak classifiers using the sample voting integration strategy. Using collective wisdom, we can get rid of bad decisions, giving the integrated model a much better generalization ability. Extensive experiments on three datasets show that BCMR can significantly improve cross-media retrieval accuracy. [ABSTRACT FROM AUTHOR]

Femtocells are the feasible solutions to extend the network coverage of indoor users and to enhance the network capacity in long-term evolution advanced (LTE-A)-based 5G networks. However, the femtocell base station shares the same frequency spectrum of microcell base station in unplanned manner. Hence, interference mitigation is a crucial problem in densely deployed femtocell environment and it is more severe with the deployment of femtocells in LTE-A network. In this paper, a modified dirty paper coding is proposed for interference mitigation along with the optimization of feedback bits using natural inspired meta-heuristic firefly algorithm. The proposed meta-heuristic algorithm reduces the interference by periodically unicasting the channel state information. Since the bandwidth of feedback system is limited, it is optimized in such a way that it does not affect the performance of the system. As compared to the conventional zero-forcing pre-coding, the proposed modified dirty paper coding along with firefly algorithm scheme offers improved sum rate of 70% and 64% with increase in the number of feedback bits and number of users, respectively. [ABSTRACT FROM AUTHOR]

The main purpose of identifying and locating the retina vessels is to specify from the fundus image the various tissues of the vascular structure, which can be wide or tight. The classification of vessels in the retinal image often confronts several challenges, such as the low contrast accompanying the fundus image, the inhomogeneity of the background lighting, and the noise. Moreover, fuzzy c-means (FCM) is one of the most frequently used algorithms for medical image segmentation due to its effectiveness. Hence, many FCM method derivatives have been developed to improve their noise robustness and time-consuming. This paper aims to analyze the performance of some improved FCM algorithms to recommend the best ones for the segmentation of retinal blood vessels. Eight derivatives of FCM algorithm are detained in this study: FCM, EnFCM, SFCM, FGFCM, FRFCM, DSFCM_N, FCM_SICM and SSFCA. The performance analysis is conducted from three viewpoints: noise robustness, blood vessels segmentation performance, and time-consuming. At first, the noise robustness of improved FCM clustering algorithms is evaluated using a synthetic image degraded by various types and levels of noise. Then, the ability of the selected algorithms to segment retinal blood vessels is assessed based on images from the DRIVE and STARE databases after a pre-processing phase. Finally, the time consumption of each algorithm is measured. The experiments demonstrate that the FRFCM and DSFCM_N algorithms achieve better results in terms of noise robustness and blood vessels segmentation. Regarding the running time, the FRFCM algorithm requires less time than other algorithms in the segmentation of retinal images. The results of this study are extensively discussed, and some suggestions are proposed at the end of this paper. [ABSTRACT FROM AUTHOR]

Transportation has always been an indispensable part of people's lives. With the growth of the total population, the number of trips continues to increase, traffic pressure continues to increase, and road traffic accidents are increasing every year. With the continuous development of artificial intelligence technology, the emergence of intelligent transportation systems has greatly promoted the development and commercialization of irreversible intelligent transportation systems. This paper aims to study the road vehicle recognition algorithm in safety-assisted driving based on artificial intelligence. The research topic of this paper is the research of road vehicle recognition algorithm in safety-assisted driving. The research focus is on lane detection algorithm and vehicle detection algorithm. The concept of lane line detection is proposed, and the image preprocessing method is introduced. The experimental results in this paper show that when the road environment is complex, the vehicle itself will cause misjudgments, but the overall correct rate is over 92.5%. [ABSTRACT FROM AUTHOR]

Manifold fitting is a manifold verification technique for data with noise and manifold structures. By extracting the expected manifold structure, the reliability of the data manifold hypothesis can be determined, and the true structure of the data without noise can conform to a manifold. This paper proposes a manifold fitting algorithm for the variable-scale spectral graph theory and estimates the deviation of the manifold structure caused by noise. Considering the scale variations in frequency-domain analysis based on spectral graph theory, details of the data under the effect of small scale and characteristics of data shape under the effect of large scale are highlighted. This study uses the average calculation and meanshift method to obtain two types of mean vectors from each neighborhood, which are essential in suppressing noise and maintaining shape, respectively. Therefore, manifold fitting is carried out from two aspects, specifically weakening noise and characterizing the manifold shape. To obtain a closer estimate of the deviation caused by noise to the manifold structure, this study also estimates the neighborhood distribution of data under the effect of medium scale, obtains the covariance information of each neighborhood, and uses the variance information to estimate the manifold structure deviations. [ABSTRACT FROM AUTHOR]

It is well recognized that human societies and their problem-solving capabilities have evolved much faster than biological evolution. The inspiration from human behaviors, knowledge exchange, and transformation has given rise to a new evolutionary computation paradigm. Multiple research endeavors have been reported in the literature inspired by the diverse aspects of human societies with corresponding terminologies to describe the algorithm. These endeavors have resulted in piles of algorithms, worded differently but with more or less similar underlying mechanisms causing immense confusion for a new reader. This paper presents a generalized framework for these socio-inspired evolutionary algorithms (SIEAs) or socio-inspired meta-heuristic algorithms. A survey of various SIEAs is provided to highlight the working of these algorithms on a common framework, their variations and improved versions proposed in the literature, and their applications in various fields of search and optimization. The algorithmic description and the comparison of each SIEA with the general framework enable a clearer understanding of the similarities and differences between these methodologies. Efforts have been made to provide an extensive list of references with due context. This paper could become an excellent reference as a starting point for anyone interested in this fascinating field of research. [ABSTRACT FROM AUTHOR]

In practical applications of graph theory, indeterminacy factors always appear in graphs. Uncertain random graph was proposed via chance theory, in which some edges exist with degrees in probability measure and others exist with degrees in uncertain measure. This paper discusses the contributions of edges for connectivity of an uncertain random graph and proposes concepts about significance of edges, according to which edges are classified. In addition, this paper presents algorithms for calculating connectivity index and significance of edges of an uncertain random graph. Examples are given to illustrate algorithms and methods. [ABSTRACT FROM AUTHOR]

Reconstruction of cross-cut shredded text documents (RCCSTD) has important applications for information security and judicial evidence collection. The traditional method of manual construction is a very time-consuming task, so the use of computer-assisted efficient reconstruction is a crucial research topic. Fragment consensus information extraction and fragment pair compatibility measurement are two fundamental processes in RCCSTD. Due to the limitations of the existing classical methods of these two steps, only documents with specific structures or characteristics can be spliced, and pairing error is larger when the cutting is more fine-grained. In order to reconstruct the fragments more effectively, this paper improves the extraction method for consensus information and constructs a new global pairwise compatibility measurement model based on the extreme learning machine algorithm. The purpose of the algorithm's design is to exploit all available information and computationally suggest matches to increase the algorithm's ability to discriminate between data in various complex situations, then find the best neighbor of each fragment for splicing according to pairwise compatibility. The overall performance of our approach in several practical experiments is illustrated. The results indicate that the matching accuracy of the proposed algorithm is better than that of the previously published classical algorithms and still ensures a higher matching accuracy in the noisy datasets, which can provide a feasible method for RCCSTD intelligent systems in real scenarios. [ABSTRACT FROM AUTHOR]

This paper designs a maritime anomaly detection algorithm based on a support vector machine (SVM) that considers the spatiotemporal and motion features of trajectories. Since trajectories are two-dimensional, it is difficult to present their motion features. To accurately describe trajectory features, a novel trajectory feature extraction method based on statistical theory is proposed in this paper. This method maps trajectories onto a high-dimensional space, which can account for both the spatiotemporal features and motion features of the trajectories. With the proposed feature extraction method, the density-based spatial clustering of applications with noise algorithm is employed to recognize vessel traffic patterns by simultaneously considering the spatiotemporal and motion features. Then, an improved SVM is designed by employing a weighted hybrid kernel function and differential operator to detect anomalous behaviours from recognized vessel traffic patterns that include the spatiotemporal and motion characteristics. Compared with standard SVM, it can adaptively determine the optimal kernel function according to sample set. Finally, a numerical example based on automatic identification system data from the waters off Chengshan Jiao is fulfilled to verify the proposed algorithm effectiveness and accuracy. [ABSTRACT FROM AUTHOR]

Semi-stream join is an emerging research problem in the domain of near-real-time data warehousing. A semi-stream join is basically a join between a fast stream (S) and a slow disk-based relation (R). In the modern era of technology, huge amounts of data are being generated swiftly on a daily basis which needs to be instantly analyzed for making successful business decisions. Keeping this in mind, a famous algorithm called CACHEJOIN (Cache Join) was proposed. The limitation of the CACHEJOIN algorithm is that it does not deal with the frequently changing trends in a stream data efficiently. To overcome this limitation, in this paper, we propose a TinyLFU-CACHEJOIN algorithm, a modified version of the original CACHEJOIN algorithm, which is designed to enhance the performance of a CACHEJOIN algorithm. TinyLFU-CACHEJOIN employs an intelligent strategy which keeps only those records of R in the cache that have a high hit rate in S. This mechanism of TinyLFU-CACHEJOIN allows it to deal with the sudden and abrupt trend changes in S. We developed a cost model for our TinyLFU-CACHEJOIN algorithm and proved it empirically. We also assessed the performance of our proposed TinyLFU-CACHEJOIN algorithm with the existing CACHEJOIN algorithm on a skewed synthetic dataset. The experiments proved that TinyLFU-CACHEJOIN algorithm significantly outperforms the CACHEJOIN algorithm. [ABSTRACT FROM AUTHOR]

Basketball is one of the three ball games, loved by all sports enthusiasts, and shooting is the most basic means of scoring. In order to improve the scoring rate of basketball shooting, the teaching objective program and method program are combined to form an efficient teaching algorithm with comprehensive characteristics, which can improve the effect of basketball shooting training. In this paper, we design reasonable auxiliary teaching methods for basketball shooting, so that students can improve their scoring rate in daily basketball training and reduce injury when improving shooting skills. Our evaluation suggests that increasing the emphasis on the training of core muscle group is beneficial to enhance the control of core muscle group of athletes and promote the improvement of their shooting percentage. For people, without the basis of foundation and has a certain movement intermittent time, longer training time while shorter core training time were observed. Moreover, for physical quality, excellent and experienced player should be aimed at training projects, which strictly controls the intermittent time, while fully stimulates the core muscles, in order to achieve the training targets and complete the training results. We conclude that it is feasible and effective to design an auxiliary teaching algorithm for basketball shooting on the basis of object decomposition program. Furthermore, the proposed algorithm is approximately 11.23% better than the other methods in terms of higher scores. [ABSTRACT FROM AUTHOR]

The performance of target detection algorithms for manual features is gradually saturated, and the development of target detection has stagnated. Computer vision is a discipline that studies how to use computers to replace the human eye and use visual information and visual algorithms to automatically detect, recognize, and track target objects. Target detection has also been fully developed as a basis for solving advanced vision tasks such as target tracking, instance segmentation, image understanding, and behavior recognition. The theories of artificial intelligence and deep learning has gradually developed, making convolutional neural networks a hot spot of attention in the field of computer vision and image processing. Based on the work in the field of target detection, this paper proposes a target detection algorithm based on the prior knowledge of the impact area of the tennis ball. Our experiments and evaluations show that the algorithm can achieve higher detection accuracy and faster detection speed. Moreover, the proposed algorithm accurately tracks and judge the impact area of tennis sports, and improve the accuracy and reliability of the tennis impact. [ABSTRACT FROM AUTHOR]

Human facial analysis (HFA) has recently become an attractive topic for computer vision research due to technological progress and mobile applications. HFA explores several issues as gender recognition (GR), facial expression, age, and race recognition for automatically understanding social life. This study explores HFA from the angle of recognizing a person's gender from their face. Several hard challenges are provoked, such as illumination, occlusion, facial emotions, quality, and angle of capture by cameras, making gender recognition more difficult for machines. The Archimedes optimization algorithm (AOA) was recently designed as a metaheuristic-based population optimization method, inspired by the Archimedes theory's physical notion. Compared to other swarm algorithms in the realm of optimization, this method promotes a good balance between exploration and exploitation. The convergence area is increased By incorporating extra data into the solution, such as volume and density. Because of the preceding benefits of AOA and the fact that it has not been used to choose the best area of the face, we propose utilizing a wrapper feature selection technique, which is a real motivation in the field of computer vision and machine learning. The paper's primary purpose is to automatically determine the optimal face area using AOA to recognize the gender of a human person categorized by two classes (Men and women). In this paper, the facial image is divided into several subregions (blocks), where each area provides a vector of characteristics using one method from handcrafted techniques as the local binary pattern (LBP), histogram-oriented gradient (HOG), or gray-level co-occurrence matrix (GLCM). Two experiments assess the proposed method (AOA): The first employs two benchmarking datasets: the Georgia Tech Face dataset (GT) and the Brazilian FEI dataset. The second experiment represents a more challenging large dataset that uses Gallagher's uncontrolled dataset. The experimental results show the good performance of AOA compared to other recent and competitive optimizers for all datasets. In terms of accuracy, the AOA-based LBP outperforms the state-of-the-art deep convolutional neural network (CNN) with 96.08% for the Gallagher's dataset. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is among the best evolutionary algorithms for global optimization. However, the basic DE has several shortcomings, like the slow convergence speed, and it is more likely to be stuck at local optima. Additionally, DE's performance is sensitive to its mutation strategies and control parameters for mutation and crossover. In this scope, we present in this paper three mechanisms to overcome DE limitations. First, two novel mutations called DE/mean-current/2 and DE/best-mean-current/2 are proposed and integrated in the DE algorithm, and they have both exploration ability and exploitation trend. On the other hand, to avoid being trapped in local minima of hard functions, a new exploration operator has been proposed called Weibull flight based on the Weibull distribution. Finally, new adapted control parameters based on the Weibull distribution are integrated. These parameters contribute to the optimization process by adjusting mutation scale and alleviating the parameter setting problem often encountered in various metaheuristics. The efficacy of the proposed algorithms called meanDE, MDEW, AMDE, and AMDEW is validated through intensive experimentations using classical tests, some challenging tests, the CEC2017, CEC2020, the most recent CEC2022, four constraint engineering problems, and the data clustering problem. Moreover, comparisons with several popular, recent, and high-performance optimization algorithms show a high effectiveness of the proposed algorithms in locating the optimal or near-optimal solutions with higher efficiency. The experiments clearly indicate the effectiveness of the new mutations compared to the standard DE mutations. Moreover, the proposed Weibull flight has a great capacity to deal with the hard composition functions of CEC benchmarks. Finally, the use of adapted control parameters for the mutation scale helps overcome the parameter setting problem commonly encountered in various metaheuristics. [ABSTRACT FROM AUTHOR]

This paper presents a multi-unmanned aerial vehicle (UAV)-assisted mobile edge computing system, where multiple UAVs are used to serve mobile users. We aim to minimize the overall energy consumption of the system by planning the trajectories of UAVs. To plan the trajectories of UAVs, we need to consider the deployment of hovering points (HPs) of UAVs, their association with UAVs, and their order for each UAV. Therefore, the problem is very complicated, as it is non-convex, nonlinear, NP-hard, and mixed-integer. To solve the problem, this paper proposed an evolutionary trajectory planning algorithm (ETPA), which comprises four phases. In the first phase, a variable-length GA is adopted to update the deployments of HPs for UAVs. Accordingly, redundant HPs are removed by the remove operator. Subsequently, a differential evolution clustering algorithm is adopted to cluster HPs into different clusters without knowing the number of HPs in advance. Finally, a GA is proposed to construct the order of HPs for UAVs. The experimental results on a set of eight instances show that the proposed ETPA outperforms other compared algorithms in terms of the energy consumption of the system. [ABSTRACT FROM AUTHOR]

In this paper, an improved particle swarm optimization (IPSO) algorithm is proposed to solve the technician routing and scheduling problem (TRSP). The TRSP consists of the assignment of technicians into teams, the assignment of teams to tasks, the construction of routes, and the selection of the day on which a service is provided by considering the proficiency level of workers and the proficiency requirement of the task. The paper considers the planning horizon as a multi-period covering 5 days, which further increases the complexity of the problem. Then a task can be fulfilled in any one of 5 days. The IPSO algorithm includes a particle swarm optimization (PSO) algorithm and one neighborhood operator. One neighborhood operator is used to avoid the local solution trap since the global best solution found by PSO is falling into a local solution trap. Further, the proposed algorithm's performance is experimentally compared with the branch-and-cut algorithm for the solution of the TRSP, on the benchmark instances generated from the literature. The computational results show that IPSO provides better solutions considering the branch-and-cut algorithm within reasonable computing time. [ABSTRACT FROM AUTHOR]

In this paper, septic B-spline approach with linearization and without linearization technique with the help of Butcher's fifth-order Runge–Kutta scheme is implemented to obtain solitary wave solutions of the modified regularized long wave equation. The error norms L 2 and L ∞ and the three invariants I 1 , I 2 and I 3 are computed on three tests to confirm about the efficiency and accuracy of the proposed methods. The stability analysis is performed using Von-Neumann technique on the linearized septic B-spline approach. The results of the present work are compared with the analytical and existing solutions. [ABSTRACT FROM AUTHOR]

Cooperation search algorithm (CSA) is a new metaheuristic algorithm inspired from the team cooperation behaviors in modern enterprises and is characterized by fast convergence. However, for complex multimodal problems, it may get trapped into local optima and suffer from premature convergence for the shortcoming of population updating guided only by leading individuals. In this paper, the issue of low convergence efficiency and convergence accuracy of the CSA algorithm on complex multimodal problems is dramatically alleviated by integrating the mutation and crossover operators in DE algorithm. Experimental results demonstrate the better performance of CCSA on convergence speed and accuracy as compared to other existing optimizers. Furthermore, aiming at the problem that there is no universal approach for the multi-degree reduction in Ball Bézier surfaces under different interpolation constrains, we propose a new method to solve this problem by introducing metaheuristic methods, where the change of interpolation constrains is treated as the change of decision variables. The modeling examples show that the proposed method is effective and easy to implement under different interpolation constrains, which can achieve the multi-degree reduction in Ball Bézier surfaces at one time and can simplify the degree reduction procedure significantly. [ABSTRACT FROM AUTHOR]

This paper presents a new way of selecting an initialisation for the k -modes algorithm that allows for a notion of game theoretic fairness that classic initialisations, namely those by Huang and Cao, do not. Our new method utilises the hospital-resident assignment problem to find the set of initial cluster centroids which we compare with two classical initialisation methods for k -modes: the original presented by Huang and the next most popular method of Cao and co-authors. To highlight the merits of our proposed method, two stages of analysis are presented. It is demonstrated that the proposed method is often able to offer computational speed-up of the order of 50 % . Improved clustering, in terms of a commonly used cost-function, was witnessed in several cases and can be of the order of 10 % , particularly for more complex datasets. [ABSTRACT FROM AUTHOR]

General type-2 fuzzy logic systems have received wide concerns in current academic subject. Type-reduction is the kernel module for the systems. This paper shows the interpretations for the beginning of Karnik–Mendel (KM) algorithms. According to the famous numerical integration technique, the weighting approaches of enhanced Karnik–Mendel (EKM) algorithms are put forward. Then, the sensible beginning weighted enhanced Karnik–Mendel (SBWEKM) algorithms are put forward to perform the centroid type-reduction. Compared with the EKM algorithms, WEKM algorithms and SBEKM algorithms, this approach helps to improve both the absolute errors and convergence speeds as shown in four computer simulation experiments. [ABSTRACT FROM AUTHOR]

Academic institutions may be edging towards a global uncertainty, recession, and a string of financial difficulties. An effective course timetabling is one of managerial strategies to optimise the operating costs and resource utilisation. This paper presents the first application of firefly algorithm (FA) and its modifications and hybridisations for solving real-world course timetabling problem. A novel bi-objective firefly algorithm (BOFA) with Pareto dominance approach was developed for optimising the operating costs and resource utilisation. Random key technique was applied for adapting the continuous firefly movement to solve discrete timetabling problem. Five constructive heuristics were additionally embedded into the BOFA for initialising feasible timetables. Computational experiments were sequentially conducted using eleven problem instances obtained from a collaborating university. It was found that the proposed hybridisation outperformed particle swarm optimisation, the classical FA, and modified FA (MFA), in terms of the quality of the timetables obtained, computational times and convergent speed. The proposed hybrid bi-objective FA (HBOFA) yielded better Pareto frontiers than the conventional BOFA with less computational times for all problem instances. [ABSTRACT FROM AUTHOR]

Hyperparameter optimization plays a significant role in the overall performance of machine learning algorithms. However, the computational cost of algorithm evaluation can be extremely high for complex algorithm or large dataset. In this paper, we propose a model-based reinforcement learning with experience variable and meta-learning optimization method to speed up the training process of hyperparameter optimization. Specifically, an RL agent is employed to select hyperparameters and treat the k-fold cross-validation result as a reward signal to update the agent. To guide the agent's policy update, we design an embedding representation called "experience variable" and dynamically update it during the training process. Besides, we employ a predictive model to predict the performance of machine learning algorithm with the selected hyperparameters and limit the model rollout in short horizon to reduce the impact of the inaccuracy of the model. Finally, we use the meta-learning technique to pre-train the model for fast adapting to a new task. To prove the advantages of our method, we conduct experiments on 25 real HPO tasks and the experimental results show that with the limited computational resources, the proposed method outperforms the state-of-the-art Bayesian methods and evolution method. [ABSTRACT FROM AUTHOR]