Global Fitness Function Research Articles

A remote health monitoring framework for heart disease and diabetes prediction using advanced artificial intelligence model

Remote health monitoring frameworks gained significant attention due to their real intervention and treatment standards. The proposed work intends to design an artificial intelligence (AI) based remote health monitoring framework for predicting heart disease and diabetes from the given medical datasets. In this framework, the smart devices are used to gather the health information of patients, and the obtained information is integrated together by using different nodes that includes the detecting node, visualization node, and prognostic node. Then, at that point, the health care dataset preprocessing is performed to standardize the characteristics by recognizing the missing qualities and taking out the unessential characteristics. Consequently, the unified levy modeled crow search optimization (ULMCSO) algorithm is employed to select the optimal features based on the global fitness function, which helps increase the accuracy and reduce the training time of the classifier. Finally, the probabilistic guided naïve distribution (PGND) based classification model is utilized for predicting the label as to whether normal or disease affected. During an evaluation, two different datasets, such as PIMA and Hungarian, are used to validate and compare the results of the proposed model by using various performance measures.

An Industry Framework for Remote Health Monitoring using Machine Learning Models to Predict a Disease

Remote health monitoring frameworks gained significant attention due to their real intervention and treatment standards. Most conventional works object to developing remote monitoring frameworks for identifying the disease at the earlier stages for an appropriate diagnosis. Still, it faced the problems with complexity in operations, increased cost of resources, misprediction results, which requires more time consumption for data gathering, and reduced convergence rate. Hence, the proposed work intends to design a machine learning based remote health monitoring framework for predicting heart disease and diabetes from the given medical datasets. In this framework, the Industry based smart devices are used to gather the health information of patients, and the obtained information is integrated together by using different nodes that includes the detecting node, visualization node, and prognostic node. Then, the medical dataset preprocessing is performed to normalize the attributes by identifying the missing values and eliminating the irrelevant qualities. Consequently, the Unified Levy Modeled Crow Search Optimization (U-CSO) algorithm is employed to select the optimal features based on the global fitness function, which helps increase the accuracy and reduce the training time of the classifier. Finally, the Most Probabilistic Guided Naïve Distribution (MP-ND) based classification model is utilized for predicting the label as to whether normal or disease affected. During an evaluation, two different datasets, such as PIMA and Hungarian, are used to validate and compare the results of the proposed model by using various performance measures. A Patients' health status can be monitored remotely for disease detection and proper diagnosis.

Three-Dimensional Force Sensor Based on Fiber Bragg Grating for Medical Puncture Robot

In medical puncture robots, visible light, infrared and ultrasound images are currently used to guide punctures. The lack of information about the interaction forces between the puncture needle and soft tissue in different directions during the puncture process can easily lead to soft tissue being damaged. The current three-dimensional force sensors are large and can only be mounted on the base of the puncture needle, which does not allow for easy integration. Moreover, the force transfer to the base introduces various disturbing forces and the measurement accuracy is low. To reduce the risk of soft tissue being damaged and to enhance the intelligent control strategy of the puncture robot, this paper designs a three-dimensional force sensor based on fiber Bragg gratings. The sensor is very small and can be integrated into the back end of the puncture needle to accurately measure the interaction forces between the puncture needle and the soft tissue in different directions. The puncture needle wall is designed with notched bending of a multilayer continuous beam, which can increase the sensitivity of axial stiffness, while maintaining the sensitivity of the sensor to lateral bending and torsion, and also reduce the crosstalk between the axial and lateral forces. The finite element method is used to optimize its structural parameters, and a BP neural network based on the global optimal fitness function is proposed to solve the decoupling problem between the three-dimensional forces, which effectively improves the detection accuracy of the force sensor. The experimental results show that the measurement error of the sensor is less than 1.5%, which can accurately measure the interaction force between the puncture needle and the soft tissue and improve the safety of the puncture process.

The Application of Art Therapy Based on Particle Swarm Optimization Method for Preschool Children’s Mental Health

Art provides a starting point for children to understand works of art. In the initial stage of life, the development of vision and movement is particularly important, and painting helps coordinate the work of these systems. Art dispels the mystery of art and helps children connect with art classics, thus understanding the language and structure of art. In particle swarm optimization, the inertia weight is constant, but with the increase of iteration times, the inertia weight may become larger or smaller, so it tends to search globally or locally. This paper first analyzes what art therapy is from the basic concepts of art and art education and the stage characteristics of children’s painting and then studies the application of art therapy based on particle swarm optimization in preschool children’s art teaching. Finding the global optimal fitness function in the research process not only simplifies the computational complexity but also approximates the maximum coverage effect provided by the current network. The results show that applying particle swarm optimization algorithm to the research of mental health treatment of preschool children can establish an effective evaluation program in the practical research of children’s art treatment.

Universal Evolutionary Model for Periodical Species

Real-world examples of periodical species range from cicadas, whose life cycles are large prime numbers, like 13 or 17, to bamboos, whose periods are large multiples of small primes, like 40 or even 120. The periodicity is caused by interaction of species, be it a predator-prey relationship, symbiosis, commensalism, or competition exclusion principle. We propose a simple mathematical model, which explains and models all those principles, including listed extremal cases. This rather universal, qualitative model is based on the concept of a local fitness function, where a randomly chosen new period is selected if the value of the global fitness function of the species increases. Arithmetically speaking, the different interactions are related to only four principles: given a couple of integer periods either (1) their greatest common divisor is one, (2) one of the periods is prime, (3) both periods are equal, or (4) one period is an integer multiple of the other.

Achieving task allocation in swarm intelligence with bi-objective embodied evolution

In this paper, we seek to achieve task allocation in swarm intelligence using an embodied evolutionary framework, which aims to generate divergent and specialized behaviors among a swarm of agents in an online and self-organized manner. In our considered scenario, specialization is encouraged through a bi-objective composite fitness function for the genomes, which is the weighted sum of a local and a global fitness function. The former depends only on the behavior of an agent itself, while the latter depends on the effectiveness of cooperation among all nearby agents. We have tested two existing variants of embodied evolution on this scenario and compared their performances against those of an individual random walk baseline algorithm. We have found out that those two embodied evolutionary algorithms have good performances at the extreme cases of weight configurations, but are not adequate when the two objective functions interact. We thus propose a novel bi-objective embodied evolutionary algorithm, which handles the aforementioned scenario by controlling the proportion of specialized behaviors via a dynamic reproductive isolation mechanism. Its performances are compared against those of other considered algorithms, as well as the theoretical Pareto frontier produced by NSGA-II.

Evolutionary Divide-and-Conquer Algorithm for Virus Spreading Control Over Networks

The control of virus spreading over complex networks with a limited budget has attracted much attention but remains challenging. This article aims at addressing the combinatorial, discrete resource allocation problems (RAPs) in virus spreading control. To meet the challenges of increasing network scales and improve the solving efficiency, an evolutionary divide-and-conquer algorithm is proposed, namely, a coevolutionary algorithm with network-community-based decomposition (NCD-CEA). It is characterized by the community-based dividing technique and cooperative coevolution conquering thought. First, to reduce the time complexity, NCD-CEA divides a network into multiple communities by a modified community detection method such that the most relevant variables in the solution space are clustered together. The problem and the global swarm are subsequently decomposed into subproblems and subswarms with low-dimensional embeddings. Second, to obtain high-quality solutions, an alternative evolutionary approach is designed by promoting the evolution of subswarms and the global swarm, in turn, with subsolutions evaluated by local fitness functions and global solutions evaluated by a global fitness function. Extensive experiments on different networks show that NCD-CEA has a competitive performance in solving RAPs. This article advances toward controlling virus spreading over large-scale networks.

Comprehensive modelling, analysis and optimization of furan resin-based moulding sand system with sawdust as an additive

Scarcity of high-cost silica sand, casting defect such as hot tear in hard moulds and casting ejection problem after solidification are the key industrial problems. Sawdust is a by-product of wood working industries, and economic disposal of sawdust in these industries is a growing concern to the wood industries. The present work utilized sawdust as an additive in preparing mould cavity for casting applications. Sand mould properties such as compression strength (CS), mould hardness (MH), gas evolution (GE), permeability (P) and collapsibility (CP) will have good impact on the quality of castings. The effect of variables, namely quantity of resin, hardener, sawdust and setting time, on no-bake furan-bonded sand system is studied in the present research work. The experiments are conducted as per design of experiments, and the data are used to investigate the effect of individual and combined parametric contributions towards responses and establish nonlinear input–output relationships. All nonlinear regression models (that is, input–output relationships) are found to be statistically adequate. The input–output relations are analysed and presented for each of the response with the help of surface plots. Further, the models are found to predict the output close to the experiment (target value). The grand average value in predicting responses is found to be equal to 5.03%. The multi-objective optimization of responses with conflicting nature (minimize: GE and CP; maximize: CS, P and MH) is carried out with the help of global fitness function values determined using genetic algorithm, particle swarm optimization, teacher–learner-based optimization and JAYA algorithms. The optimized values of process parameters that resulted in best set of responses are found to be equal to 60 min, 2.01%, 0.6% and 0.93% for setting time, quantity of resin, hardener and sawdust, respectively. Two automobile coupling parts are cast by pouring molten aluminium into the mould cavity with the optimized and non-optimum sand mould conditions. Further, these two cast components are tested for their quality characteristics, such as surface finish, yield strength, hardness, density and secondary dendrite arm spacing. It has been observed that the quality characteristics of castings produced in mould with optimized parameters are found to be much better.

How to ‘find’ an error minimized genetic code: neutral emergence as an alternative to direct Darwinian selection for evolutionary optimization

Error minimization (EM) of the standard genetic code (SGC) refers to the assignment of amino acids to codons in such a way that the deleterious impact of mutations is reduced. The SGC is nearly optimal for the property of EM, compared to randomly generated codes, and prompts the question of how the property arose. Brute force searching of alternative genetic codes is unlikely to have occurred, given the high number of alternative codes. Therefore, a heuristic search of ‘code space’, the space of alternative codes, would have been necessary. Uncovering the nature of this heuristic search is key to understanding the evolution of the genetic code, and consequently the origin of life. Scenarios that rely on direct selection for the property of EM require codon reassignments to sample code space, but these are problematic mechanistically. Alternatively, it has been shown that EM may have emerged in a neutral fashion as a byproduct of the process of genetic code expansion. In this scenario, similar amino acids are added to similar codons via the gene duplication of tRNAs and aminoacyl-tRNA synthetases. Mimicking this process via simulation indeed produces high levels of EM in the resulting genetic codes. These observations imply that optimization has occurred by an alternative to direct selection, commonly viewed as the only form of evolutionary optimization followed in nature. I propose that the neutral emergence of EM produced by code expansion is a genetic algorithm but unlike direct selection, the local selection criterion (amino acid and codon similarity) is distant from the global fitness function (EM), leading to the emergent optimization of EM. By presenting this counter example I clarify how evolutionary optimization in biological systems is not restricted to direct selection, and emphasize that additional processes may lead to the production of beneficial traits, via ‘non-Darwinian optimization’.

Tackling learning intractability through topological organization and regulation of cortical networks.

A key challenge in evolving control systems for robots using neural networks is training tractability. Evolving monolithic fixed topology neural networks is shown to be intractable with limited supervision in high dimensional search spaces. Common strategies to overcome this limitation are to provide more supervision by encouraging particular solution strategies, manually decomposing the task and segmenting the search space and network. These strategies require a supervisor with domain knowledge and may not be feasible for difficult tasks where novel concepts are required. The alternate strategy is to use self-organized task decomposition to solve difficult tasks with limited supervision. The artificial neural tissue (ANT) approach presented here uses self-organized task decomposition to solve tasks. ANT inspired by neurobiology combines standard neural networks with a novel wireless signaling scheme modeling chemical diffusion of neurotransmitters. These chemicals are used to dynamically activate and inhibit wired network of neurons using a coarse-coding framework. Using only a global fitness function that does not encourage a predefined solution, modular networks of neurons are shown to self-organize and perform task decomposition. This approach solves the sign-following task found to be intractable with conventional fixed and variable topology networks. In this paper, key attributes of the ANT architecture that perform self-organized task decomposition are shown. The architecture is robust and scalable to number of neurons, synaptic connections, and initialization parameters.

High-Dimensional Waveform Inversion With Cooperative Coevolutionary Differential Evolution Algorithm

In this letter, an improved differential evolution (DE) for high-dimensional waveform inversion is proposed. In conventional evolutionary algorithms, an individual is treated as a whole, and all its variables (genes) are evaluated with a uniform fitness function. This evaluation criterion is not effective for a high-dimensional individual. Therefore, for high-dimensional waveform inversion, we incorporate the decomposition strategy of cooperative coevolution into DE to decompose the individual into some subcomponents. Another novel feature that we introduce is a local fitness function for each subcomponent, and a new mutation operator is designed to guide the mutation direction of each subcomponent with the corresponding local fitness value. Coevolution among different subcomponents is realized in the selection operation with the global fitness function. Many experiments have been carried out to evaluate the performance of this new algorithm. The results clearly show that, for high-dimensional waveform inversion, this algorithm is effective and performs better than some other methods. Finally, the new method has been applied to real seismic data.