Search Phase Research Articles

Anxiety, depression and acromegaly: a systematic review.

scientific literature highlights risk factors linked to the onset of psychopathology in different medical pathological contexts. Acromegaly is a rare condition, particularly noteworthy due to the associated clinical psychological features. This research aimed at understanding the main psychopathological outcomes related to acromegaly, with particular emphasis to anxiety and depression. In January 2024, in line with PRISMA guidelines, a systematic search based on PubMed, Scopus, Web of Science and PsycInfo was conducted to detect studies considering anxiety, depression and alexithymia in patients suffering from acromegaly. The Keywords used for the search phase were "Acromegaly" AND "Depression" OR "Anxiety" OR "Alexithymia". Fifty-five studies were eligible. Anxiety and depression were significantly present in patients with acromegaly, with prevalence rates variable based on disease status and psycho-diagnostic instruments. None of the included studies reported alexithymia in patients with acromegaly. No significant difference was found regarding anxiety and depressive symptoms in patients with acromegaly in comparison with patients suffering from different pituitary diseases and chronic conditions. Anxiety and depression were associated with lower perceived HR-QoL, presence of comorbidity, joint issues, delayed diagnosis, disease duration and body image concerns. Anxiety and depression may be encountered in patients with acromegaly, impacting HR-QoL and the course of the disease. This systematic review suggests that a deeper evaluation of clinical psychological features in patients suffering from acromegaly is needed. Particularly, the early detection of clinical psychological symptoms may lead to multi-integrate interventions promoting individuals' well-being and a better HR-QoL.

Hybrid secure routing and monitoring mechanisms in IoT-based wireless sensor networks using Egret-Harris optimization

ABSTRACT Security is the primary concern when creating security protocols for wireless sensor networks (WSN), which motivates many academics to find security solutions that effectively provide a few benefits, such as low consumption of electricity, flexible communication, and low cost. A few restrictions still remain, including the inability to exactly select the expected cluster, the sensor nodes’ constrained functionality, and their poor efficiency. Thus, Egret-Harris optimization for hybrid secure routing and monitoring mechanisms in IoT-based WSNs (EHO optimized routing protocol in WSN) was introduced in this research. The created EHO algorithm combines the search, fitness function, and hunting phase features from Harris hawks and egrets to determine the best solution among all practical solutions while ensuring safe data transfer from the cluster heads (CHs) to the Base station (BS). Specifically, the EHO-enabled clustering is applied to the suggested model to efficiently choose the ideal group of CHs. Additionally, the EHO algorithm assists in choosing the best possible routes with minimal distance and less delay for facilitating energy-efficient transmission. With 100 nodes analyzed, the suggested EHO-WSN approach without any attacks achieved 22 alive nodes, a delay of 0.10 ms, a normalized energy of 0.346J, and a throughput of 0.64 bps, respectively. Additionally, in the presence of a Sybil attack, the suggested EHO-WSN technique achieves 14, 0.214 J, 0.010 ms, and 0.512 bps for an analysis involving 100 nodes. Compared to previous methods, the suggested EHO-WSN model without attack achieves a delay of 0.07 ms, a throughput of 0.30 bps, an energy of 0.374 J, and 37 alive nodes for 200 node evaluation. For 200 nodes under examination, the EHO-WSN technique yields superior results of attaining 12 alive nodes, a delay of 0.072 ms, a throughput of 0.181 bps, and a normalized energy of 0.330 J even in the presence of the Sybil attack and exceeded other traditional techniques.

Effectiveness of Self-Management Programs Among Athletes With Patellofemoral Pain Syndrome: Protocol for a Systematic Review.

Patellofemoral pain syndrome is a highly prevalent overuse knee injury in athletic populations associated with pain and functional limitations, exacerbated by activities such as running, pivoting, cycling, and jumping. Self-management programs empowering athletes to take an active role in controlling their symptoms for chronic musculoskeletal conditions such as patellofemoral pain syndrome have grown in popularity. However, the efficacy of self-management programs specifically for athletes with patellofemoral pain syndrome is unclear due to limited and heterogeneous evidence. The systematic review study will evaluate the effectiveness of self-management programs on pain and function, in athletes with patellofemoral pain syndrome. PubMed/MEDLINE, Cochrane Library, CINAHL (Cumulative Index to Nursing and Allied Health Literature), and PsycINFO databases will be systematically searched using terms related to "patellofemoral pain syndrome," "self-management," and "athletes." Interventional studies that are randomized and nonrandomized controlled trials will be included, comparing self-management programs to other treatments or control conditions among athletes with patellofemoral pain syndrome. Four reviewers will independently screen studies, extract data using the COVIDENCE software, and assess the quality of the study and evidence using the Pedro scale of risk of bias tool and GRADE approach, respectively. If feasible, a meta-analysis will be performed using the RevMan (version 5.4; the Cochrane Collaboration). The systematic review is currently in the search phase, with the authors refining search strings for the selected databases. The final search strings are expected to be ready by March 2024, and the review is projected to be completed by July 2024. This systematic review protocol outlines a rigorous methodology for evaluating the effectiveness of self-management programs among athletes with patellofemoral pain syndrome. The findings will inform clinical practice and guide the development of tailored interventions to optimize outcomes for athletes with patellofemoral pain syndrome. PROSPERO CRD42023492746; https://tinyurl.com/c5jze9ca. PRR1-10.2196/58340.

An Improved Cuckoo Search Algorithm and Its Application in Robot Path Planning

This manuscript introduces an improved Cuckoo Search (CS) algorithm, known as BASCS, designed to address the inherent limitations of CS, including insufficient search space coverage, premature convergence, low search accuracy, and slow search speed. The proposed improvements encompass four main areas: the integration of tent chaotic mapping and random migration in population initialization to reduce the impact of random errors, the guidance of Levy flight by the directional determination strategy of the Beetle Antennae Search (BAS) algorithm during the global search phase to improve search accuracy and convergence speed, the adoption of the Sine Cosine Algorithm for local exploitation in later iterations to enhance local optimization and accuracy, and the adaptive adjustment of the step-size factor and elimination probability throughout the iterative process to convergence. The performance of BASCS is validated through ablation experiments on 10 benchmark functions, comparative experiments with the original CS and its four variants, and application to a robot path planning problem. The results demonstrate that BASCS achieves higher convergence accuracy and exhibits faster convergence speed and superior practical applicability compared to other algorithms.

Adaptive collective responses to local stimuli in anonymous dynamic networks

We develop a framework for self-induced phase changes in programmable matter in which a collection of agents with limited computational and communication capabilities can collectively perform appropriate global tasks in response to local stimuli that dynamically appear and disappear. Agents are represented by vertices in a dynamic graph G whose edge set changes over time, and stimuli are placed adversarially on the vertices of G where each agent is only capable of recognizing a co-located stimulus. Agents communicate via token passing along edges to alert other agents to transition to an Aware state when stimuli are present and an Unaware state when the stimuli disappear. We present an Adaptive Stimuli Algorithm that can handle arbitrary adversarial stimulus dynamics, while an adversary (or the agents themselves) reconfigures the connections (edges) of G over time in a controlled way. This algorithm can be used to solve the foraging problem on reconfigurable graphs where, in addition to food sources (stimuli) being discovered, removed, or shifted arbitrarily, we would like the agents to consistently self-organize, using only local interactions, such that if the food remains in a position long enough, the agents transition to a gather phase in which many collectively form a single large component with small perimeter around the food. Alternatively, if no food source has existed recently, the agents should undergo a self-induced collective phase change and switch to a search phase in which they distribute themselves randomly throughout the graph to search for food. Unlike previous approaches to foraging, this process is indefinitely repeatable, withstanding competing broadcast waves of state transition that may interfere with each other. Like a physical phase change, such as the ferromagnetic models underlying the gather and search algorithms used for foraging, microscopic changes in the environment trigger these macroscopic, system-wide transitions as agents share information and respond locally to get the desired collective response.

What do we know about the role of neoadjuvant targeted therapy in early-stage EGFR-mutant and ALK-fused non-small cell lung cancer?-a narrative review of the current literature.

The standard first-line treatment for patients with advanced non-small cell lung cancer (NSCLC) harbouring epidermal growth factor receptor (EGFR) mutations or anaplastic lymphoma kinase (ALK) fusions is targeted therapy using tyrosine kinase inhibitors (TKIs). However, data are still lacking on the use of TKIs as a neoadjuvant or induction approach. Therefore, this narrative review aims to summarize the current knowledge on resectable EGFR-mutant and ALK-fused NSCLC regarding available perioperative treatment regimens and off-label neoadjuvant use of targeted therapy. The relevant literature was identified by using PubMed and ClinicalTrials.gov (last search phase June 2024) and was restricted to English language. Peer-reviewed manuscripts but also conference abstracts that did not undergo peer-review were included. Patients with EGFR-mutations and ALK-fusions have typically been excluded from available phase III perioperative immunotherapy trials due to lower efficacy and higher toxicity of immunotherapy in those patients. In the adjuvant setting, recent evidence from the phase III ALINA and ADAURA trials demonstrated efficacy and safety of targeted therapy in resected ALK-fused and EGFR-mutant NSCLC. However, to date there is no approval for the use of TKIs as neoadjuvant or induction therapy in those patients. We have therefore identified a number of case series and phase II trials using targeted therapy in resectable EGFR-mutant and ALK-fused NSCLC. Current evidence suggests that targeted therapies might be effective in patients with resectable EGFR-mutant and ALK-positive NSCLC, but ongoing trials will need to provide further evidence on the safety and efficacy of perioperative TKI therapy.

Solving the yard crane scheduling problem with dynamic assignment of input/output points

This paper introduces an automatic yard crane scheduling problem with additional assignments of input/output (I/O) points at the block. I/O points are the buffer between the block and the rest of the terminal and containers are relocated from the block to the I/O points or vice-versa. The crane schedule therefore not only considers movements, storage and retrieval of containers, but must also be coordinated with the release and due times of containers at the I/O points, which are also limited and need to be assigned during the scheduling process. This results in a complex, but at the same time, more realistic problem. Two GRASP (Greedy Randomized Adaptive Search Procedure) heuristic approaches are proposed along with improvements to the solution construction and local search phases that are specifically tailored for this problem. The proposed algorithms are statistically calibrated and comprehensive experiments are designed to assess the performance of the method, including validation across small, medium, and large size instances. The experimental results show that the proposed methods are effective and competitive when compared to existing approaches. The GRASP manages to obtain an optimality average rate of 71.25% for small instances and finds 461 new best solutions in the 480 medium and large instances, with up to 200 containers, when compared to existing approaches for this problem.

Covid-19 pandemic trouble? An exploratory study on German acquisitions in Japan

PurposeThis study investigated how the Covid-19 pandemic impacted cross-border acquisitions. Though literature suggests that cross-border investments decreased during the pandemic, there is little conclusive evidence on specific characteristics in the execution of particular acquisitions during such times. We applied the case study format to conduct our investigation by (1) providing a classification of influences on cross-border procedures and (2) highlighting critical characteristics during three phases of acquisitions, namely, search, negotiation and integration periods.Design/methodology/approachThe grounded theory approach was applied to three in-depth case studies of German companies that acquired Japanese targets during the Covid-19 pandemic. Data were supplemented by information received through additional case studies of German subsidiaries in Japan and interviews with consultants.FindingsFirms had already intended to acquire their respective targets, with their decisions having been made prior to the Covid-19 pandemic. Thus, the pandemic had no impact on target selection in the case firms. Owing to travel restrictions, information exchange was limited which inevitably led to higher usage of digitalization. While several barriers led to delays in negotiations, prevailing mutual trust and assistance from consultants helped to reduce difficulties. During the integration period, we found delays in synergy creation and increases in remote communication. Nevertheless, the digital workflow improved the efficiency.Originality/valueOur study provides novel insights into the execution of cross-border acquisitions impacted by the Covid-19 pandemic. We discuss new implications for mergers and acquisitions (M&A) research and practice for the post-pandemic era, focusing on German firms acquiring Japanese targets.

Inverse design of bistable composite laminates with switching tunneling method for global optimization

Bistability enables adaptive designs with tunable deflections for applications including morphing wings, robotic grippers, and consumer products. Composite laminates may be designed to exhibit bistability due to pre-strains that develop during the processing of the polymer matrix, enabling fast reconfiguration between two stable shapes. Unfortunately, designing bistable laminates is challenging because of their highly nonlinear behavior. Here, we propose the Switching Tunneling Method to address this challenge by alternating between gradient-based local minimization and tunneling search phases, with the enhancement of objective expression switching to improve numerical conditioning. Results demonstrate high effectiveness compared to existing optimizers; the Switching Tunneling Method achieves a 99% success rate in finding all energy minima across general composite layups. Additionally, our method facilitates the inverse design of variable pre-strain fields, enabling bioinspired, positive Gaussian curvatures, which are not possible with conventional pre-strain laminates. Validations through both finite element analysis and 3D printed samples confirm the optimal designs.

Mechanism of homology search expansion during recombinational DNA break repair in Saccharomyces cerevisiae

Homology search is a central step of DNA double-strand break (DSB) repair by homologous recombination (HR). How it operates in cells remains elusive. We developed a Hi-C-based methodology to map single-stranded DNA (ssDNA) contacts genome-wide in S.cerevisiae, which revealed two main homology search phases. Initial search conducted by short Rad51-ssDNA nucleoprotein filaments (NPFs) is confined in cis by cohesin-mediated chromatin loop folding. Progressive growth of stiff NPFs enables exploration of distant genomic sites. Long-range resection drives this transition from local to genome-wide search by increasing the probability of assembling extensive NPFs. DSB end-tethering promotes coordinated search by opposite NPFs. Finally, an autonomous genetic element on chromosome III engages the NPF, which stimulates homology search in its vicinity. This work reveals the mechanism of the progressive expansion of homology searchthat isorchestrated by chromatin organizers, long-range resection, end-tethering,andspecialized genetic elements andthatexploits the stiff NPF structure conferred by Rad51 oligomerization.

Adaptive multi-stage evolutionary search for constrained multi-objective optimization

In this paper, we propose a multi-stage evolutionary framework with adaptive selection (MSEFAS) for efficiently handling constrained multi-objective optimization problems (CMOPs). MSEFAS has two stages of optimization in its early phase of evolutionary search: one stage that encourages promising infeasible solutions to approach the feasible region and increases diversity, and the other stage that enables the population to span large infeasible regions and accelerates convergence. To adaptively determine the execution order of these two stages in the early process, MSEFAS treats the optimization stage with higher validity of selected solutions as the first stage and the other as the second one. In addition, at the late phase of evolutionary search, MSEFAS introduces a third stage to efficiently handle the various characteristics of CMOPs by considering the relationship between the constrained Pareto fronts (CPF) and unconstrained Pareto fronts. We compare the proposed framework with eleven state-of-the-art constrained multi-objective evolutionary algorithms on 56 benchmark CMOPs. Our results demonstrate the effectiveness of the proposed framework in handling a wide range of CMOPs, showcasing its potential for solving complex optimization problems.

Parameter optimization of the simulated moving bed system based on the IMOSCSO algorithm

AbstractThe optimization of operating parameters for the simulated moving bed (SMB) is complex. A parameter optimization method for the SMB system was proposed based on the improved multi‐objective sand cat swarm optimization (IMOSCSO) algorithm. The multi‐objective sand cat swarm optimization (MOSCSO) algorithm integrated the update and selection mechanism of the repository in the multi‐objective algorithm. Three strategies were proposed to improve the traditional MOSCSO algorithm for increased population diversity, global search capability, and convergence speed. First, the cubic chaotic map was used to initialize the population, which improved the uniformity of the population distribution. Second, including a variable spiral search strategy in the prey search phase enabled the sand cat swarm to explore more search paths to adjust its position. Third, the convergence speed was enhanced by incorporating the alert mechanism of the sparrow search algorithm. The improved algorithm was tested with standard test functions. The IMOSCSO algorithm outperformed other algorithms in terms of convergence and accuracy. Finally, the IMOSCSO algorithm optimized the system parameters of the SMB, demonstrating its practical applications.

Application of Local Search Particle Swarm Optimization Based on the Beetle Antennae Search Algorithm in Parameter Optimization

Intelligent control algorithms have been extensively utilized for adaptive controller parameter adjustment. While the Particle Swarm Optimization (PSO) algorithm has several issues: slow convergence speed requiring a large number of iterations, a tendency to get trapped in local optima, and difficulty escaping from them. It is also sensitive to the distribution of the solution space, where uneven distribution can lead to inefficient contraction. On the other hand, the Beetle Antennae Search (BAS) algorithm is robust, precise, and has strong global search capabilities. However, its limitation lies in focusing on a single individual. As the number of iterations increases, the step size decays, causing it to get stuck in local extrema and preventing escape. Although setting a fixed or larger initial step size can avoid this, it results in poor stability. The PSO algorithm, which targets a population, can help the BAS algorithm increase diversity and address its deficiencies. Conversely, the characteristics of the BAS algorithm can aid the PSO algorithm in finding the optimal solution early in the optimization process, accelerating convergence. Therefore, considering the combination of BAS and PSO algorithms can leverage their respective advantages and enhance overall algorithm performance. This paper proposes an improved algorithm, W-K-BSO, which integrates the Beetle Antennae Search strategy into the local search phase of PSO. By leveraging chaotic mapping, the algorithm enhances population diversity and accelerates convergence speed. Additionally, the adoption of linearly decreasing inertia weight enhances algorithm performance, while the coordinated control of the contraction factor and inertia weight regulates global and local optimization performance. Furthermore, the influence of beetle antennae position increments on particles is incorporated, along with the establishment of new velocity update rules. Simulation experiments conducted on nine benchmark functions demonstrate that the W-K-BSO algorithm consistently exhibits strong optimization capabilities. It significantly improves the ability to escape local optima, convergence precision, and algorithm stability across various dimensions, with enhancements ranging from 7 to 9 orders of magnitude compared to the BAS algorithm. Application of the W-K-BSO algorithm to PID optimization for the Pointing and Tracking System (PTS) reduced system stabilization time by 28.5%, confirming the algorithm’s superiority and competitiveness.

Relationship between Neuromarketing and Online Purchasing Decisions of Peruvian Consumers in the Post-Pandemic Era

The pandemic has accelerated the transition toward e-commerce, highlighting the necessity of understanding how neuromarketing affects online purchasing decisions. This quantitative study, based on a correlational and cross-sectional design, explored the relationship between neuromarketing and purchasing decisions in virtual stores in northern Peru. Using a 40-item questionnaire, data from 498 adults in La Libertad and Lambayeque were collected. The analyses revealed significant correlations between cognition attention and the information search (rho=0.58) and alternative evaluation phases (rho=0.53), between emotion and the postpurchase stage (rho=0.54), and between memory and the recognition of need (rho=0.57), in addition to a strong direct association between cognition attention and the overall influence of neuromarketing on purchasing decisions (rho=0.72). These findings underscore the considerable impact of sensory and emotional experiences generated by neuromarketing on the critical stages of the purchasing behavior of Peruvian digital consumers, pointing to its relevance in enhancing customer satisfaction and loyalty in the digital age. In conclusion, various components of neuromarketing, specifically linked to the generation of pleasant and memorable sensory experiences, have a significant influence on the search, evaluation, and postpurchase satisfaction behaviors of virtual consumers in the region studied. Received: 3 March 2024 / Accepted: 13 June 2024 / Published: 02 July 2024

Latency-Aware Neural Architecture Performance Predictor With Query-to-Tier Technique

Neural Architecture Search (NAS) is a powerful tool for automating effective image and video processing DNN designing. The ranking of the accuracy has been advocated to design an efficient performance predictor for NAS. The previous contrastive method solves the ranking problem by comparing pairs of architectures and predicting their relative performance. However, it only focuses on the rankings between the two involved architectures and neglects the overall quality distributions of the search space, which may suffer generalization issues. On the contrary, we propose to let the performance predictor concentrate on the global quality level of specific architecture, and learn the tier embeddings of the whole search space automatically with learnable queries. The proposed method, dubbed as Neural Architecture Ranker with Query-to-Tier technique (NARQ2T), explores the quality tiers of the search space globally and classifies each individual to the tier they belong to. Thus, the predictor gains knowledge of the performance distributions of the search space which helps to generalize its ranking ability to the datasets more easily. Thanks to the encoder-decoder design, our method is able to predict the latency of the searched model without deteriorating the performance prediction. Meanwhile, the global quality distribution facilitates the search phase by directly sampling candidates according to the statistics of quality tiers, which is free of training a search algorithm, e.g., Reinforcement Learning or Evolutionary Algorithm, thus it simplifies the NAS pipeline and saves the computational overheads. The proposed NARQ2T achieves state-of-the-art performance on two widely used datasets for NAS research. Moreover, extensive experiments have validated the efficacy of the designed method.

Joint Dual Feature Distillation and Gradient Progressive Pruning for BERT compression

The increasing size of pre-trained language models has led to a growing interest in model compression. Pruning and distillation are the primary methods employed to compress these models. Existing pruning and distillation methods are effective in maintaining model accuracy and reducing its size. However, they come with limitations. For instance, pruning is often suboptimal and biased by transforming it into a continuous optimization problem. Distillation relies primarily on one-to-one layer mappings for knowledge transfer, which leads to underutilization of the rich knowledge in teacher. Therefore, we propose a method of joint pruning and distillation for automatic pruning of pre-trained language models. Specifically, we first propose Gradient Progressive Pruning (GPP), which achieves a smooth transition of indicator vector values from real to binary by progressively converging the values of unimportant units’ indicator vectors to zero before the end of the search phase. This effectively overcomes the limitations of traditional pruning methods while supporting compression with higher sparsity. In addition, we propose the Dual Feature Distillation (DFD). DFD adaptively globally fuses teacher features and locally fuses student features, and then uses the dual features of global teacher features and local student features for knowledge distillation. This realizes a “preview-review” mechanism that can better extract useful information from multi-level teacher information and transfer it to student. Comparative experiments on the GLUE benchmark dataset and ablation experiments indicate that our method outperforms other state-of-the-art methods.

Fast Filter Pruning via Coarse-to-Fine Neural Architecture Search and Contrastive Knowledge Transfer.

Filter pruning is the most representative technique for lightweighting convolutional neural networks (CNNs). In general, filter pruning consists of the pruning and fine-tuning phases, and both still require a considerable computational cost. So, to increase the usability of CNNs, filter pruning itself needs to be lightweighted. For this purpose, we propose a coarse-to-fine neural architecture search (NAS) algorithm and a fine-tuning structure based on contrastive knowledge transfer (CKT). First, candidates of subnetworks are coarsely searched by a filter importance scoring (FIS) technique, and then the best subnetwork is obtained by a fine search based on NAS-based pruning. The proposed pruning algorithm does not require a supernet and adopts a computationally efficient search process, so it can create a pruned network with higher performance at a lower cost than the existing NAS-based search algorithms. Next, a memory bank is configured to store the information of interim subnetworks, i.e., by-products of the above-mentioned subnetwork search phase. Finally, the fine-tuning phase delivers the information of the memory bank through a CKT algorithm. Thanks to the proposed fine-tuning algorithm, the pruned network accomplishes high performance and fast convergence speed because it can take clear guidance from the memory bank. Experiments on various datasets and models prove that the proposed method has a significant speed efficiency with reasonable performance leakage over the state-of-the-art (SOTA) models. For example, the proposed method pruned the ResNet-50 trained on Imagenet-2012 up to 40.01% with no accuracy loss. Also, since the computational cost amounts to only 210 GPU hours, the proposed method is computationally more efficient than SOTA techniques. The source code is publicly available at https://github.com/sseung0703/FFP.

A Hybrid Preference Interaction Mechanism for Multi-Satellite Imaging Dynamic Mission Planning

The existing multi-satellite dynamic mission planning system hardly satisfies the requirements of fast response time and high mission benefit in highly dynamic situations. In the meantime, a reasonable decision-maker preference mechanism is an additional challenge for multi-satellite imaging dynamic mission planning based on user preferences (MSDMPUP). Therefore, this study proposes the hybrid preference interaction mechanism and knowledge transfer strategy for the multi-objective evolutionary algorithm (HPIM–KTSMOEA). Firstly, an MSDMPUP model based on a task rolling window is constructed to achieve timely updating of the target task importance degree through the simultaneous application of periodic triggering and event triggering methods. Secondly, the hybrid preference interaction mechanism is constructed to plan according to the satellite controller’s preference-based commands in different phases of the optimal search of the mission planning scheme to effectively respond to the dynamic changes in the environment. Finally, a knowledge transfer strategy for the multi-objective evolutionary algorithm is proposed to accelerate population convergence in new environments based on knowledge transfer according to environmental variability. Simulation experiments verify the effectiveness and stability of the method in processing MSDMPUP. This study found that the HPIM–KTSMOEA algorithm has high task benefit, short response time, and high task completion when processing MSDMPUP.

Automatic time series forecasting model design based on pruning

Automatic Time Series Forecasting (TSF) model design which aims to help users to efficiently design forecasting models for the given time series data scenarios, is a novel research topic to be urgently solved. In this paper, we propose AutoTS algorithm trying to utilize the existing design skills and design efficient search methods to effectively solve this problem. In AutoTS, we extract effective design experience from the existing TSF works. We allow the effective combination of design experience from different sources, to create an effective search space containing a variety of TSF models to support different TSF tasks. Considering the huge search space, in AutoTS, we propose a two-stage pruning strategy to reduce the search difficulty and improve the search efficiency. Specifically, at the beginning of the search phase, we apply the vertical pruning method to quickly optimize each module of the TSF model. In the middle of the search, we turn to apply the horizontal pruning method to filter out less effective options for each module according to the learned experience and optimize modules. Moreover, in AutoTS, we introduce the knowledge graph to reveal associations between module options. We make full use of this relational information to learn higher-level features of each module option, to further improve the search quality. Experimental results show that AutoTS is well-suited for the TSF area. On some general datasets, our algorithms discover the assembled model with better performance. Additionally, compared to existing NAS algorithms, they exhibit higher search efficiency and greater search potential than the manually designed ones.

Optimal deployment of private 5G multi-access edge computing systems at smart factories: Using hybrid crow search algorithm

In smart factories, an increasing number of mobile intelligent devices are deployed to meet the growing demands for flexible manufacturing. These devices, equipped with various sensors, synchronize a substantial amount of data with cloud servers for real-time monitoring and control. Fifth generation mobile networks (5G) combined with multi-access edge computing (MEC) provide the capabilities for multiple connections, large-scale data transmission, and efficient response times, becoming the mainstream network architectures for the needs. Additionally, with the emerging trend of energy-saving, recent works have addressed energy consumption as a significant cost factor. However, most previous studies tended to focus solely on cloud servers, neglecting energy consumption at edge computing and underestimating the impact of heterogeneous mobile device. As the scope of smart factory networks continues to expand, the challenges become increasingly critical. This study presents a private 5G MEC system architecture tailored for smart panel factories. The concerned problem is formulated as an integer programming model, which determines deployment locations and quantities for MEC servers and 5G small cells, including picocells and femtocells, so as to minimize the overall deployment costs while meeting the constraints of connectivity, capacity, latency, coverage, serviceability, and energy consumption. Since the edge server deployment problem is NP-hard, this study further proposes a hybrid metaheuristic algorithm, CSAVNS, which combines the strengths of crow search algorithm (CSA) and variable neighborhood search (VNS). In the global search phase of CSA, the VNS local search is introduced to enhance the algorithm's capabilities. Experimental analysis demonstrates that the proposed CSAVNS outperforms other algorithms in terms of solution solving capabilities.