Search results

Nature’s evolution has shaped intelligent behaviors in creatures like insects and birds, inspiring the field of Swarm Intelligence. Researchers have developed bio-inspired algorithms to address complex optimization problems efficiently. These algorithms strike a balance between computational efficiency and solution optimality, attracting significant attention across domains.

Bio-inspired optimization techniques for feature engineering and its applications are systematically reviewed with chief objective of assessing statistical influence and significance of “Bio-inspired optimization”-based computational models by referring to vast research literature published between year 2015 and 2022.

The Scopus and Web of Science databases were explored for review with focus on parameters such as country-wise publications, keyword occurrences and citations per year. Springer and IEEE emerge as the most creative publishers, with indicative prominent and superior journals, namely, PLoS ONE, Neural Computing and Applications, Lecture Notes in Computer Science and IEEE Transactions. The “National Natural Science Foundation” of China and the “Ministry of Electronics and Information Technology” of India lead in funding projects in this area. China, India and Germany stand out as leaders in publications related to bio-inspired algorithms for feature engineering research.

The review findings integrate various bio-inspired algorithm selection techniques over a diverse spectrum of optimization techniques. Anti colony optimization contributes to decentralized and cooperative search strategies, bee colony optimization (BCO) improves collaborative decision-making, particle swarm optimization leads to exploration-exploitation balance and bio-inspired algorithms offer a range of nature-inspired heuristics.

The purpose of this paper is to propose a whole set of methods for underwater target detection, because most underwater objects have small samples, low quality underwater images problems such as detail loss, low contrast and color distortion, and verify the feasibility of the proposed methods through experiments.

The improved RGHS algorithm to enhance the original underwater target image is proposed, and then the YOLOv4 deep learning network for underwater small sample targets detection is improved based on the combination of traditional data expansion method and Mosaic algorithm, expanding the feature extraction capability with SPP (Spatial Pyramid Pooling) module after each feature extraction layer to extract richer feature information.

The experimental results, using the official dataset, reveal a 3.5% increase in average detection accuracy for three types of underwater biological targets compared to the traditional YOLOv4 algorithm. In underwater robot application testing, the proposed method achieves an impressive 94.73% average detection accuracy for the three types of underwater biological targets.

Underwater target detection is an important task for underwater robot application. However, most underwater targets have the characteristics of small samples, and the detection of small sample targets is a comprehensive problem because it is affected by the quality of underwater images. This paper provides a whole set of methods to solve the problems, which is of great significance to the application of underwater robot.

With the rapid growth of the domestic lending industry, assessing whether the borrower of each loan is at risk of default is a pressing issue for financial institutions. Although there are some models that can handle such problems well, there are still some shortcomings in some aspects. The purpose of this paper is to improve the accuracy of credit assessment models.

In this paper, three different stages are used to improve the classification performance of LSTM, so that financial institutions can more accurately identify borrowers at risk of default. The first approach is to use the K-Means-SMOTE algorithm to eliminate the imbalance within the class. In the second step, ResNet is used for feature extraction, and then two-layer LSTM is used for learning to strengthen the ability of neural networks to mine and utilize deep information. Finally, the model performance is improved by using the IDWPSO algorithm for optimization when debugging the neural network.

On two unbalanced datasets (category ratios of 700:1 and 3:1 respectively), the multi-stage improved model was compared with ten other models using accuracy, precision, specificity, recall, G-measure, F-measure and the nonparametric Wilcoxon test. It was demonstrated that the multi-stage improved model showed a more significant advantage in evaluating the imbalanced credit dataset.

In this paper, the parameters of the ResNet-LSTM hybrid neural network, which can fully mine and utilize the deep information, are tuned by an innovative intelligent optimization algorithm to strengthen the classification performance of the model.

In order to estimate the value of semi-automated subject indexing in operative library catalogues, the study aimed to investigate five different automated implementations of an open source software package on a large set of Swedish union catalogue metadata records, with Dewey Decimal Classification (DDC) as the target classification system. It also aimed to contribute to the body of research on aboutness and related challenges in automated subject indexing and evaluation.

On a sample of over 230,000 records with close to 12,000 distinct DDC classes, an open source tool Annif, developed by the National Library of Finland, was applied in the following implementations: lexical algorithm, support vector classifier, fastText, Omikuji Bonsai and an ensemble approach combing the former four. A qualitative study involving two senior catalogue librarians and three students of library and information studies was also conducted to investigate the value and inter-rater agreement of automatically assigned classes, on a sample of 60 records.

The best results were achieved using the ensemble approach that achieved 66.82% accuracy on the three-digit DDC classification task. The qualitative study confirmed earlier studies reporting low inter-rater agreement but also pointed to the potential value of automatically assigned classes as additional access points in information retrieval.

The paper presents an extensive study of automated classification in an operative library catalogue, accompanied by a qualitative study of automated classes. It demonstrates the value of applying semi-automated indexing in operative information retrieval systems.

Bayesian cubature (BC) has emerged to be one of most competitive approach for estimating the multi-dimensional integral especially when the integrand is expensive to evaluate, and alternative acquisition functions, such as the Posterior Variance Contribution (PVC) function, have been developed for adaptive experiment design of the integration points. However, those sequential design strategies also prevent BC from being implemented in a parallel scheme. Therefore, this paper aims at developing a parallelized adaptive BC method to further improve the computational efficiency.

By theoretically examining the multimodal behavior of the PVC function, it is concluded that the multiple local maxima all have important contribution to the integration accuracy as can be selected as design points, providing a practical way for parallelization of the adaptive BC. Inspired by the above finding, four multimodal optimization algorithms, including one newly developed in this work, are then introduced for finding multiple local maxima of the PVC function in one run, and further for parallel implementation of the adaptive BC.

The superiority of the parallel schemes and the performance of the four multimodal optimization algorithms are then demonstrated and compared with the k-means clustering method by using two numerical benchmarks and two engineering examples.

Multimodal behavior of acquisition function for BC is comprehensively investigated. All the local maxima of the acquisition function contribute to adaptive BC accuracy. Parallelization of adaptive BC is realized with four multimodal optimization methods.

Metaheuristic algorithms based on biology, evolutionary theory and physical principles, have been widely developed for complex global optimization. This paper aims to present a new hybrid optimization algorithm that combines the characteristics of biogeography-based optimization (BBO), invasive weed optimization (IWO) and genetic algorithms (GAs).

The significant difference between the new algorithm and original optimizers is a periodic selection scheme for offspring. The selection criterion is a function of cyclic discharge and the fitness of populations. It differs from traditional optimization methods where the elite always gains advantages. With this method, fitter populations may still be rejected, while poorer ones might be likely retained. The selection scheme is applied to help escape from local optima and maintain solution diversity.

The efficiency of the proposed method is tested on 13 high-dimensional, nonlinear benchmark functions and a homogenous slope stability problem. The results of the benchmark function show that the new method performs well in terms of accuracy and solution diversity. The algorithm converges with a magnitude of 10-4, compared to 102 in BBO and 10-2 in IWO. In the slope stability problem, the safety factor acquired by the analogy of slope erosion (ASE) is closer to the recommended value.

This paper introduces a periodic selection strategy and constructs a hybrid optimizer, which enhances the global exploration capacity of metaheuristic algorithms.

Traditional case-adaptation methods have poor accuracy, low efficiency and limited applicability, which cannot meet the needs of knowledge users. To address the shortcomings of the existing research in the industry, this paper proposes a case-adaptation optimization algorithm to support the effective application of tacit knowledge resources.

The attribute simplification algorithm based on the forward search strategy in the neighborhood decision information system is implemented to realize the vertical dimensionality reduction of the case base, and the fuzzy C-mean (FCM) clustering algorithm based on the simulated annealing genetic algorithm (SAGA) is implemented to compress the case base horizontally with multiple decision classes. Then, the subspace K-nearest neighbors (KNN) algorithm is used to induce the decision rules for the set of adapted cases to complete the optimization of the adaptation model.

The findings suggest the rapid enrichment of data, information and tacit knowledge in the field of practice has led to low efficiency and low utilization of knowledge dissemination, and this algorithm can effectively alleviate the problems of users falling into “knowledge disorientation” in the era of the knowledge economy.

This study provides a model with case knowledge that meets users’ needs, thereby effectively improving the application of the tacit knowledge in the explicit case base and the problem-solving efficiency of knowledge users.

The adaptation model can serve as a stable and efficient prediction model to make predictions for the effects of the many logistics and e-commerce enterprises' plans.

This study designs a multi-decision class case-adaptation optimization study based on forward attribute selection strategy-neighborhood rough sets (FASS-NRS) and simulated annealing genetic algorithm-fuzzy C-means (SAGA-FCM) for tacit knowledgeable exogenous cases. By effectively organizing and adjusting tacit knowledge resources, knowledge service organizations can maintain their competitive advantages. The algorithm models established in this study develop theoretical directions for a multi-decision class case-adaptation optimization study of tacit knowledge.

Prefabricated building has been widely applied in the construction industry all over the world, which can significantly reduce labor consumption and improve construction efficiency compared with conventional approaches. During the construction of prefabricated buildings, the overall efficiency largely depends on the lifting sequence and path of each prefabricated component. To improve the efficiency and safety of the lifting process, this study proposes a framework for automatically optimizing the lifting path of prefabricated building components using building information modeling (BIM), improved 3D-A* and a physic-informed genetic algorithm (GA).

Firstly, the industry foundation class (IFC) schema for prefabricated buildings is established to enrich the semantic information of BIM. After extracting corresponding component attributes from BIM, the models of typical prefabricated components and their slings are simplified. Further, the slings and elements’ rotations are considered to build a safety bounding box. Secondly, an efficient 3D-A* is proposed for element path planning by integrating both safety factors and variable step size. Finally, an efficient GA is designed to obtain the optimal lifting sequence that satisfies physical constraints.

The proposed optimization framework is validated in a physics engine with a pilot project, which enables better understanding. The results show that the framework can intuitively and automatically generate the optimal lifting path for each type of prefabricated building component. Compared with traditional algorithms, the improved path planning algorithm significantly reduces the number of nodes computed by 91.48%, resulting in a notable decrease in search time by 75.68%.

In this study, a prefabricated component path planning framework based on the improved A* algorithm and GA is proposed for the first time. In addition, this study proposes a safety-bounding box that considers the effects of torsion and slinging of components during lifting. The semantic information of IFC for component lifting is enriched by taking into account lifting data such as binding positions, lifting methods, lifting angles and lifting offsets.

A new triage method, MBCE (Medical Bio Social Ethics), is presented with social justice, bio, and medical ethics for critical resource distribution during a pandemic. Ethical triage is a complex and challenging process that requires careful consideration of medical, social, cultural, and ethical factors to guide the decision-making process and ensure fair and transparent allocation of resources. When assigning priorities to patients, a clinician would evaluate each patient’s medical condition, age, comorbidities, and prognosis, as well as their cultural and social background and ethical factors.

A statistical analysis shows no interactions among the ethical triage factors. It implies the ethical components have no moderation effect; hence, each is independent. The result also points out that medical and bioethics may have an affinity for interactions. In such cases, there seem to be some ethical factors related to bio and medical ethics that are correlated. Therefore, the triage team should be careful in evaluating patient cases. The algorithm is explained with case histories of the selected patient. A group of triage nurses and general medical practitioners assists with the triage.

The MBCE triage algorithm aims to allocate scarce resources fairly and equitably. Another ethical principle in this triage algorithm is the principle of utility. In a pandemic, the principle of utility may require prioritizing patients with a higher likelihood of survival or requiring less medical care. The research presents a sensitivity analysis of a patient’s triage score to show the algorithm’s robustness. A weighted score of ethical factors combined with an assessment of triage factors combines multiple objectives to assign a fair triage score. These distinctive features of the algorithm are reasonably easy to implement and a new direction for the unbiased triage principle.

The idea is to make decisions about distributing and using scarce medical resources. Triage algorithms raise ethical issues, such as discrimination and justice, guiding medical ethics in treating patients with terminal diseases or comorbidity. One of the main ethical principles in triage algorithms is the principle of distributive justice.

Artificial intelligence (AI) has become increasingly important and influential in reading and writing. The influx of social media digital spaces, like TikTok, has also shifted the ways multimodal composition takes place alongside AI. This study aims to argue that within spaces like TikTok, human composers must attend to the ways they write for, with and against the AI-powered algorithm.

Data collection was drawn from a larger study on #BookTok (the TikTok subcommunity for readers) that included semi-structured interviews including watching and reflecting on a TikTok they created. The authors grounded this study in critical posthumanist literacies to analyze and open code five #BookTok content creators’ interview transcripts. Using axial coding, authors collaboratively determined three overarching and entangled themes: writing for, with and against.

Findings highlight the nuanced ways #BookTokers consider the AI algorithm in their compositional choices, namely, in the ways how they want to disseminate their videos to a larger audience or more niche-focused community. Throughout the interviews, participants revealed how the AI algorithm was situated differently as both audience member, co-author and censor.

This study is grounded in critical posthumanist literacies and explores composition as a joint accomplishment between humans and machines. The authors argued that it is necessary to expand our human-centered notions of what it means to write for an audience, to co-author and to resist censorship or gatekeeping.