Search

Global warming has seriously affected the local climate characteristics of cities, resulting in the frequent occurrence of urban waterlogging with severe economic losses and casualties. Aiming to improve the effectiveness of disaster emergency management, we propose a novel emergency decision model embedding similarity algorithms of heterogeneous multi-attribute based on case-based reasoning. First, this paper establishes a multi-dimensional attribute system of urban waterlogging catastrophes cases based on the Wuli-Shili-Renli theory. Due to the heterogeneity of attributes of waterlogging cases, different algorithms to measure the attribute similarity are designed for crisp symbols, crisp numbers, interval numbers, fuzzy linguistic variables, and hesitant fuzzy linguistic term sets. Then, this paper combines the best-worst method with the maximal deviation method for a more reasonable weight allocation of attributes. Finally, the hybrid similarity between the historical and the target cases is obtained by aggregating attribute similarities via the weighted method. According to the given threshold value, a similar historical case set is built whose emergency measures are used to provide the reference for the target case. Additionally, a case of urban waterlogging emergency is conducted to demonstrate the applicability and effectiveness of the proposed model, which exploits historical experiences and retrieves the optimal scheme for the current disaster emergency with heterogeneous multi attributes. Consequently, the proposed model solves the problem of diverse data types to satisfy the needs of case presentation and retrieval. Compared with the existing model, it can better realize the multi-dimensional expression and fast matching of the cases.

This paper focuses on a new algorithm for solving optimization problems using the nature of food search behaviour of caterpillars. The paper describes how the periscopic, pheromonic and fractal search properties analogous to the caterpillars, can aid in designing a new optimization algorithm. The performance characteristics of the new method is compared using 26 standard test functions and the area under the curve of the fitness evaluations is used to validate and compare the proposed algorithms against existing related works. The proposed algorithm is found to be efficient when compared with the existing methods. The proposed algorithm is then tested on a real world problem to remove signal noise from eye gaze data, effectively.

One of the most popular population-based metaheuristic algorithms is Harris hawks optimization (HHO), which imitates the hunting mechanisms of Harris hawks in nature. Although HHO can obtain optimal solutions for specific problems, it stagnates in local optima solutions. In this paper, an improved Harris hawks optimization named ERHHO is proposed for solving global optimization problems. Firstly, we introduce tent chaotic map in the initialization stage to improve the diversity of the initialization population. Secondly, an exploration factor is proposed to optimize parameters for improving the ability of exploration. Finally, a random walk strategy is proposed to enhance the exploitation capability of HHO further and help search agent jump out the local optimal. Results from systematic experiments conducted on 23 benchmark functions and the CEC2017 test functions demonstrated that the proposed method can provide a more reliable solution than other well-known algorithms.

Dynamic economic dispatch (DED) is a multi constraint and nonlinear complex problem, which is embodied in the dynamic decision-making coupled with each other in time and space. It is generally transformed into a high-dimensional multi constraint optimization problem. In this paper, a multi Strategy firefly algorithm (MSRFA) is proposed to solve the DED problem. MSRFA puts forward three strategies through the idea of opposite learning strategy and rough data reasoning to optimize the initialization and iteration process of the algorithm, improve the convergence speed of the algorithm in medium and high dimensions, and improve the escape ability when the algorithm falls into local optimization; The performance of MSRFA is tested in the simulation experiment of DED problem. The experimental results show that MSRFA can search the optimal power generation cost and minimum load error in the experiment, which reflects MSRFA superior stability and ability to jump out of local optimization. Therefore, MSRFA is an efficient way to solve the DED problem.

This study presents the Slime Mold Algorithm (SMA) to solve the time—cost—quality trade-off problem in a construction project. The proposed SMA is a flexible and efficient algorithm in exploration and exploitation to reach the best optimal solution to process the input model’s data. This paper aims to discuss and solve the optimization problem and compare the evaluation with other algorithms such as Opposition-based Multiple Objective Differential Evolution, Non-dominated sorting genetic algorithm, Multiple objective particle swarm optimization, Multiple objective differential evolution and Chaotic initialized multiple objective differential evolution (CAMODE) to verify the efficiency and potential of the proposed algorithm. According to the analysis results, the SMA model generated a diversification measure for case studies, producing superior outcomes to those of previous algorithms.

The complexity of engineering optimization problems is increasing. Classical gradient-based optimization algorithms are a mathematical means of solving complex problems whose ability to do so is limited. Metaheuristics have become more popular than exact methods for solving optimization problems because of their simplicity and the robustness of the results that they yield. Recently, population-based bio-inspired algorithms have been demonstrated to perform favorably in solving a wide range of optimization problems. The jellyfish search optimizer (JSO) is one such bio-inspired metaheuristic algorithm, which is based on the food-finding behavior of jellyfish in the ocean. According to the literature, JSO outperforms many well-known meta-heuristics in a wide range of benchmark functions and real-world applications. JSO can also be used in conjunction with other artificial intelligence-related techniques. The success of JSO in solving diverse optimization problems motivates the present comprehensive discussion of the latest findings related to JSO. This paper reviews various issues associated with JSO, such as its inspiration, variants, and applications, and will provide the latest developments and research findings concerning JSO. The systematic review contributes to the development of modified versions and the hybridization of JSO to improve upon the original JSO and present variants, and will help researchers to develop superior metaheuristic optimization algorithms with recommendations of add-on intelligent agents.

The grasshopper optimization algorithm (GOA) is a meta-heuristic algorithm proposed in 2017 mimics the biological behavior of grasshopper swarms seeking food sources in nature for solving optimization problems. Nonetheless, some shortcomings exist in the origin GOA, and GOA global search ability is more or less insufficient and precision also needs to be further improved. Although there are many different GOA variants in the literature, the problem of inefficient and rough precision has still emerged in GOA variants. Aiming at these deficiencies, this paper develops an improved version of GOA with Levy Flight mechanism called LFGOA to alleviate the shortcomings of the origin GOA. The LFGOA algorithm achieved a more suitable balance between exploitation and exploration during searching for the most promising region. The performance of LFGOA is tested using 23 mathematical benchmark functions in comparison with the eight well-known meta-heuristic algorithms and seven real-world engineering problems. The statistical analysis and experimental results show the efficiency of LFGOA. According to obtained results, it is possible to say that the LFGOA algorithm can be a potential alternative in the solution of meta-heuristic optimization problems as it has high exploration and exploitation capabilities.

This paper introduces a new human-based metaheuristic algorithm called Sewing Training-Based Optimization (STBO), which has applications in handling optimization tasks. The fundamental inspiration of STBO is teaching the process of sewing to beginner tailors. The theory of the proposed STBO approach is described and then mathematically modeled in three phases: (i) training, (ii) imitation of the instructor’s skills, and (iii) practice. STBO performance is evaluated on fifty-two benchmark functions consisting of unimodal, high-dimensional multimodal, fixed-dimensional multimodal, and the CEC 2017 test suite. The optimization results show that STBO, with its high power of exploration and exploitation, has provided suitable solutions for benchmark functions. The performance of STBO is compared with eleven well-known metaheuristic algorithms. The simulation results show that STBO, with its high ability to balance exploration and exploitation, has provided far more competitive performance in solving benchmark functions than competitor algorithms. Finally, the implementation of STBO in solving four engineering design problems demonstrates the capability of the proposed STBO in dealing with real-world applications.

In recent years, dynamic programming and reinforcement learning theory have been widely used to solve the nonlinear control system (NCS). Among them, many achievements have been made in the construction of network model and system stability analysis, but there is little research on establishing control strategy based on the detailed requirements of control process. Spurred by this trend, this paper proposes a detail-reward mechanism (DRM) by constructing the reward function composed of the individual detail evaluation functions in order to replace the utility function in the Hamilton-Jacobi-Bellman (HJB) equation. And this method is introduced into a wider range of deep reinforcement learning algorithms to solve optimization problems in NCS. After the mathematical description of the relevant characteristics of NCS, the stability of iterative control law is proved by Lyapunov function. With the inverted pendulum system as the experiment object, the dynamic environment is designed and the reward function is established by using the DRM. Finally, three deep reinforcement learning algorithm models are designed in the dynamic environment, which are based on Deep Q-Networks, policy gradient and actor-critic. The effects of different reward functions on the experimental accuracy are compared. The experimental results show that in NCS, using the DRM to replace the utility function in the HJB equation is more in line with the detailed requirements of the designer for the whole control process. By observing the characteristics of the system, designing the reward function and selecting the appropriate deep reinforcement learning algorithm model, the optimization problem of NCS can be solved.

As one of the most popular combinatorial optimization problems, Traveling Salesman Problem (TSP) has attracted lots of attention from academia since it was proposed. Numerous meta-heuristics and heuristics have been proposed and used to solve the TSP. Although Ant Colony Optimization (ACO) is a natural TSP solving algorithm, in the process of solving it, there are also some shortcomings such as slow convergence speed and prone to fall into local optimum. Therefore, this paper proposes an improved ant colony optimization based on graph convolutional network: Graph Convolutional Network Improved Ant Colony Optimization (GCNIACO). The graph convolutional network is introduced to generate a better solution, and the better solution is converted into the pheromone on the initial path of the ACO. Thereby, the guiding effect of the pheromone concentration for the ants at the beginning of the algorithm is enhanced. In the meantime, through adaptive dynamic adjustment of the pheromone volatility factor and the introduction of the 3-opt algorithm, the algorithm's ability to jump out of the local optimum is enhanced. Finally, GCNIACO is simulated on TSP datasets and engineering application example. Comparing the optimization results with other classical algorithms, it is verified that the graph convolutional network improved ant colony optimization has better performance in obtaining the optimal solution.

The dwarf mongoose optimization (DMO) algorithm developed in 2022 was applied to solve continuous mechanical engineering design problems with a considerable balance of the exploration and exploitation phases as a metaheuristic approach. Still, the DMO is restricted in its exploitation phase, somewhat hindering the algorithm's optimal performance. In this paper, we proposed a new hybrid method called the BDMSAO, which combines the binary variants of the DMO (or BDMO) and simulated annealing (SA) algorithm. In the modelling and implementation of the hybrid BDMSAO algorithm, the BDMO is employed and used as the global search method and the simulated annealing (SA) as the local search component to enhance the limited exploitative mechanism of the BDMO. The new hybrid algorithm was evaluated using eighteen (18) UCI machine learning datasets of low and medium dimensions. The BDMSAO was also tested using three high-dimensional medical datasets to assess its robustness. The results showed the efficacy of the BDMSAO in solving challenging feature selection problems on varying datasets dimensions and its outperformance over ten other methods in the study. Specifically, the BDMSAO achieved an overall result of 61.11% in producing the highest classification accuracy possible and getting 100% accuracy on 9 of 18 datasets. It also yielded the maximum accuracy obtainable on the three high-dimensional datasets utilized while achieving competitive performance regarding the number of features selected.

Many real-world problems can be classified as multimodal optimization problems (MMOPs), which require to locate global optima as more as possible and refine the accuracy of found optima as high as possible. When dealing with MMOPs, how to divide population and obtain effective niches is a key to balance population diversity and convergence during evolution. In this paper, a self-organizing map (SOM) based differential evolution with dynamic selection strategy (SOMDE-DS) is proposed to improve the performance of differential evolution (DE) in solving MMOPs. Firstly, a SOM based method is introduced as a niching technique to divide population reasonably by using the similarity information among different individuals. Secondly, a variable neighborhood search (VNS) strategy is proposed to locate more possible optimal regions by expanding the search space. Thirdly, a dynamic selection (DS) strategy is designed to balance exploration and exploitation of the population by taking advantages of both local search strategy and global search strategy. The proposed SOMDE-DS is compared with several widely used multimodal optimization algorithms on benchmark CEC'2013. The experimental results show that SOMDE-DS is superior or competitive with the compared algorithms.

In this paper, a new stochastic optimization algorithm is introduced, called Driving Training-Based Optimization (DTBO), which mimics the human activity of driving training. The fundamental inspiration behind the DTBO design is the learning process to drive in the driving school and the training of the driving instructor. DTBO is mathematically modeled in three phases: (1) training by the driving instructor, (2) patterning of students from instructor skills, and (3) practice. The performance of DTBO in optimization is evaluated on a set of 53 standard objective functions of unimodal, high-dimensional multimodal, fixed-dimensional multimodal, and IEEE CEC2017 test functions types. The optimization results show that DTBO has been able to provide appropriate solutions to optimization problems by maintaining a proper balance between exploration and exploitation. The performance quality of DTBO is compared with the results of 11 well-known algorithms. The simulation results show that DTBO performs better compared to 11 competitor algorithms and is more efficient in optimization applications.

Alchemical language has been addressed in several lexicographical studies, historically and recently. This paper discusses the current state of the field and proposes a digital distant-reading approach to the issue of decoding alchemical

While the challenge of estimating the efficacy of therapies using observational data has received a lot of attention, little work has been done on estimating the treatment effect from interventions. In this paper, we tackle this problem by proposing an early guidance system based on a causal Bayesian network (CBN) for recommending personalized interventions. We are interested in the elderly fall prevention context. The objective is to develop a practical tool to help doctors estimate the effects of each intervention (or compound interventions) on a given patient and then choose the one that best fits each patient's health situation to reduce the risk of falling.On a real-world elderly information base, we undertake an empirical investigation for the proposed approach, which is based on a 44-node CBN. Then, we describe what is possible to achieve using state-of-the-art machine learning methods, namely Support Virtual Machine (SVM), Decision Tree (DT), and Bayesian Network (BN), and how well these methods can be used in recommending personalized interventions compared to the proposed approach.1174 elderly patients from Lille University Hospital, between January 2005 and December 2018 are included. The results reveal that none of the classifiers is significantly superior to the others, even if BN delivers somewhat better results (41.6%) and DT most often slightly lower performance (31.2%). Results also show that none of these three classifiers performs comparable to the proposed system (89.7%). The interventions recommended by the system are in agreement with the expert's judgment to a satisfactory level. The reaction of the physicians to the proposed system in its first trial version was very favorable.The study showed the efficacy and utility of the causality-based strategy in recommending tailored interventions to prevent elderly falls compared to automated learning methods that had failed to infer a solid interventional paradigm for precision medicine.

Because of the No Free Lunch (NFL) rule, we are still under the way developing new algorithms and improving the capabilities of the existed algorithms. Under consideration of the simple and steady convergence capability of the sine cosine algorithm (SCA) and the fast convergence rate of the Harris Hawk optimization (HHO) algorithms, we hereby propose a new hybridization algorithm of the SCA and HHO algorithm in this paper, called the CSCAHHO algorithm henceforth. The energy parameter is introduced to balance the exploration and exploitation procedure for individuals in the new swarm, and chaos is introduced to improve the randomness. Updating equations is redefined and combined of the equations in the SCA and HHO algorithms. Simulation experiments on 27 benchmark functions and CEC 2014 competitive functions, together with 3 engineering problems are carried out. Comparisons have been made with the original SCA, HHO, Archimedes optimization algorithm (AOA), Seagull optimization algorithm (SOA), Sooty Tern optimization algorithm (STOA), Arithmetic optimizer (AO) and Chimp optimization algorithm (ChOA). Simulation experiments on either unimodal or multimodal, benchmark or CEC2014 functions, or real engineering problems all verified the better performance of the proposed CSAHHO, such as faster convergence rate, low residual errors, and steadier capability. Matlab code of this algorithm is shared in Gitee with the following address: https://gitee.com/yuj-zhang/cscahho.

Precise customer requirements acquisition is the primary stage of product conceptual design, which plays a decisive role in product quality and innovation. However, existing customer requirements mining approaches pay attention to the offline or online customer comment feedback and there has been little quantitative analysis of customer requirements in the analogical reasoning environment. Latent and innovative customer requirements can be expressed by analogical inspiration distinctly. In response, this paper proposes a semantic analysis-driven customer requirements mining method for product conceptual design based on deep transfer learning and improved latent Dirichlet allocation (ILDA). Initially, an analogy-inspired verbal protocol analysis experiment is implemented to obtain detailed customer requirements descriptions of elevator. Then, full connection layers and a softmax layer are added to the output-end of Chinese bidirectional encoder representations from Transformers (BERT) pre-training language model. The above deep transfer model is utilized to realize the customer requirements classification among functional domain, behavioral domain and structural domain in the customer requirement descriptions of elevator by fine-tuning training. Moreover, the ILDA is adopted to mine the functional customer requirements that can represent customer intention maximally. Finally, an effective accuracy of customer requirements classification is acquired by using the BERT deep transfer model. Meanwhile, five kinds of customer requirements of elevator and corresponding keywords as well as their weight coefficients in the topic-word distribution are extracted. This work can provide a novel research perspective on customer requirements mining for product conceptual design through natural language processing.

Raven’s Progressive Matrices (RPM) is highly correlated with human intelligence, and it has been widely used to measure the abstract reasoning ability of humans. In this paper, to study the abstract reasoning capability of deep neural networks, we propose the first unsupervised learning method for solving RPM problems. Since the ground truth labels are not allowed, we design a pseudo target based on the prior constraints of the RPM formulation to approximate the ground-truth label, which effectively converts the unsupervised learning strategy into a supervised one. However, the correct answer is wrongly labelled by the pseudo target, and thus the noisy contrast will lead to inaccurate model training. To alleviate this issue, we propose to improve the model performance with negative answers. Moreover, we develop a decentralization method to adapt the feature representation to different RPM problems. Extensive experiments on three datasets demonstrate that our method even outperforms some of the supervised approaches. Our code is available at https://github.com/visiontao/ncd .

Swarm intelligence is a discipline which makes use of a number of agents for solving optimization problems by producing low cost, fast and robust solutions. The dragonfly algorithm (DA), a recently proposed swarm intelligence algorithm, is inspired by the dynamic and static swarming behaviors of dragonflies, and it has been found to have a higher performance in comparison to other swarm intelligence and evolutionary algorithms in numerous applications. There are only a few surveys about the dragonfly algorithm, and we have found that they are limited in certain aspects. Hence, in this paper, we present a more comprehensive survey about DA, its applications in various domains, and its performance as compared to other swarm intelligence algorithms. We also analyze the hybrids of DA, the methods they employ to enhance the original DA, their performance as compared to the original DA, and their limitations. Moreover, we categorize the hybrids of DA according to the type of problem that they have been applied to, their objectives, and the methods that they utilize.