Food price prediction is an important part of food supply chain management, and food price prediction is of great research significance and application value. Due to the problems of local optimization and slow learning convergence of traditional back propagation (BP) neural network, this paper proposed a BP neural network optimized by immune algorithm-particle swarm optimization hybrid algorithm (IA-PSO-BP) model for food price prediction, and selected relevant indicators. After neural network training and testing, it is found that compared with the BP neural network model optimized by particle swarm optimization algorithm (PSO-BP), the BP neural network model optimized by immune algorithm (IA-BP) and the BP neural network model, the IA-PSO-BP neural network model has the smallest prediction error and the highest accuracy. Compared with PSO-BP neural network model, IA-BP neural network model and BP neural network model, the mean absolute error (MAE) of IA-PSO-BP neural network model is reduced by 39.20%, 67.00% and 140.80%, the root mean square error (RMSE) is reduced by 39.60%, 67.50% and 141.20%, and the mean absolute percentage error (MAPE) is reduced by 0.46%, 0.79% and 1.67%. The predicted value of the IA-PSO-BP neural network model is closest to the actual value, which is more accurate to predict the actual food price. The conclusion shows that the IA-PSO-BP neural network model is a forecasting tool with higher accuracy and stronger applicability, and can be better used in food price prediction.
Currently, the rapid development of artificial intelligence, blockchain, cloud computing, big data and other “ABCD” technology has brought new momentum and new value to logistics enterprises. This paper aims to construct a digital transformation capability maturity index system for logistics enterprises, to indicate the direction of digital transformation for logistics enterprises and to promote their digital transformation process. First, the hierarchical structure of the indicator system is established: 48 indicators in total. Among them, there are 4 primary indicators, namely, digital strategic planning and organizational management capacity, infrastructure construction, business process and management digitization and digital performance, 12 secondary indicators and 32 tertiary indicators. Analytical Hierarchy Process (AHP) is applied to determine the weights of the indicators. Finally, through indicator weights analysis, relevant strategies for logistics enterprises to enhance digital transformation capability maturity are proposed, such as intelligent logistics planning algorithms should be used to improve the coordination of enterprise logistics planning, digital benefits-oriented implementation of digital transformation, top management should strongly support enterprise digital transformation, and digital security technology to strengthen the security and management and maintenance, and so on.
In this article, an in-depth analysis of research in the field of smart logistics is conducted utilizing bibliometrics and visualization tools. The study employs visualization techniques to reveal research hot spots, collaboration networks, and developmental trends within the field, as evidenced by metrics such as document counts, cooperative relationships, and keywords. Furthermore, the research employs citation network analysis and author impact assessment to explore significant research achievements and scholarly influence within the domain.
With the applications of artificial intelligence (AI) technology, graph neural networks (GNNs) have emerged and employed as a framework to handle tasks with graph-structured data. On the other hand, enhancing the efficiency of communication management system is an urgent demand to the power grid enterprises. In this paper, we propose a path recommendation framework with GNN for power communication management systems. First, historical data are collected as datasets. Second, the historical data are converted to structured form data, which are further used to construct graphs. Finally, with the proposed GNN, paths are recommended by the softmax layer, in which the one with the highest probability is the result. This scheme is verified with real data in a communication management system, where one optimal path and several candidate paths are reported in practice.
With the improvement of residents’ consumption level and urbanization level, the smart home consumption market shows a growing trend. China’s smart home consumption market is in the initial development stage, consumers’ cognition is not perfect and their willingness to purchase is still relatively low. This article elaborates on the factors influencing smart home consumption market based on consumer behavior theory, and analyzes the relationship between the factors influencing smart home consumption market. At the same time, the hierarchical structure between related factors based on ISM method to get the top-level factors is decomposed. Based on the findings of the research and analysis, it can better provide suggestions for companies to attract consumers to buy smart home products.
The fast-track project concept has become crucial in today’s competitive business landscape, driven by escalating market demands. This research aims to pinpoint success-determining factors in such projects, a quest initiated through an extensive literature review and further explored via structured interviews and questionnaires. This study unfolds in multiple phases, beginning with interviews of five experts with over 25 years of experience validating 10 critical factors: planning, scheduling, customer commitment, communication, risk management, technology adoption, project skills, material management, activity breakdown, and leadership. Subsequent phases involve quantitative approaches, including a survey with 125 respondents, predominantly Filipino professionals with substantial construction experience, and a follow-up phase with 200 Filipino respondents. These phases leverage principal component and confirmatory factor analysis, alongside machine learning techniques like K-means clustering, to analyze each success factor’s significance and perception variations. The analysis found high component loadings above 0.6 for critical success factors, confirmed by Cronbach’s Alpha values within the acceptable range of 0.6 to 0.8, but faced challenges in model fit, indicated by mixed results in Confirmatory Factor Analysis with communalities below 0.4 and inconsistent appropriate indices. The paper also looks into the application of the K-means clustering algorithm. Finally, it lays the groundwork for future research, emphasizing the need for diverse, experienced perspectives in understanding fast-track project dynamics.
Private medical institutions played a vital role in providing healthcare services to the Macao community, especially in the years leading up to 1980. This study examined the societal contribution of these institutions by analyzing inpatient data from Macao’s largest private hospital service. The key findings were as follows. Firstly, there was a consistent upward trend in the number of new inpatients (NIs) and newly discharged inpatients (NDIs), indicating a gradual increase in Kiang Wu Hospital’s impact on Macao. Secondly, Kiang Wu Hospital consistently delivered a high standard of medical care. Thirdly, the hospital’s contribution exhibited significant fluctuations over time. Lastly, the β coefficients for NI were statistically significant, whereas NDI did not show a clear convergence pattern.
With the rapid development of live-streaming and the implementation of rural revitalization, selling agricultural products via live-streaming has become an important means to achieve rural revitalization. However, the issues of quality control, supply chain, and after-sales service of the agricultural products harm the customers’ experience and the satisfaction of purchasing these products, leading to customers reducing purchase intentions. Therefore, the live-streaming platforms for supporting farmers and agricultural products have an urgent need to find the optimal strategy to increase customers’ purchase intentions. To satisfy the urgent need, this paper adopts the Bayesian learning method to model the beliefs updating of the customers’ purchase intentions for agricultural products and explores the optimal information delivery strategies for live-streaming platforms to improve customers’ purchase intentions. According to the study’s findings, in order to increase customers’ purchase intentions, live-streaming Platforms should choose different information delivery strategies for customers with different preferences. In addition, the live-streaming platforms should pay more attention to customers with high preferences, because they are more likely to purchase products and can bring greater benefits. The results of this paper could provide managerial insights to the live-streaming platforms about choosing appropriate information delivery strategies.
Health information technology can improve people’s knowledge and participation in medical decision-making. Enhance the efficiency and quality of healthcare and help reduce medical errors, costs and paperwork. Health passbook is a kind of health information technology. It can provide personal health information, manage personal medical conditions and monitor one’s overall health, which is crucial for the government to achieve sustainable development goals. Therefore, this study develops and tests a theoretical model to predict factors that influence people’s behavioral intentions to use their health passbook. It combines the extended Unified Theory of Acceptance and Use of Technology model, Task Technology Fit model and perceived enjoyment to explore the influencing factors of behavioral intention of using health passbook. In terms of questionnaire survey methods, online questionnaires and on-site face-to-face questionnaires are used to collect data. Structural equation modeling was used to analyze 1441 samples from central Taiwan. Partial least squares regression is used for hypothesis testing of the model. research shows: (1). perceived enjoyment, performance expectancy, and social influence were important antecedent factors for the public’s behavioral intentions to use health passbook; (2) . facilitating conditions have a negative impact on behavioral intentions; (3) . perceived enjoyment and tasktechnology fit were a significant determinant of both performance expectancy and effort expectancy; (4) . task characteristics and technology characteristics were key predictors of the task-technology fit. The research results and findings will help to understand the influence of the public on the intention to use the health passbook and provide relevant suggestions.
High productivity. Production line is one of the commonly used production methods in manufacturing enterprises. Improving the overall efficiency and production capacity of production line, reducing production tempo and pursuing production synchronization are more and more valued by manufacturing enterprises. Refrigerator is one of the main products of H company, which is produced by the way of production line. The investigation found that the capacity of H company’s refrigerator box is insufficient, and the production efficiency needs to be improved. Aiming at this problem, this paper takes the refrigerator box production line of H company as the research object to study its balance problem. Using 0-1 integer programming, the mathematical model is constructed with the goal of minimizing the production time, and the LINGO software is used to solve it. The production line operation elements are redistributed, and the balance rate of the production line is improved to a certain extent after optimization.
Our research focuses on the recorded data related to traffic issues around schools in the “12345” citizen service hotline of Beijing. Utilizing the Biterm Topic Model (BTM) analysis method, we conduct text mining on citizen complaints to identify six major complaint topics. The findings affirm the effectiveness of citizen complaints in identifying traffic issues around schools. Furthermore, corresponding governance measures are proposed. This study provides valuable insights and recommendations for the governance of transportation around schools in Beijing, with a focus on improving the well-being of the citizens.
In the domain of property services, traditional rule-based work order dispatching systems are challenged by issues of accuracy and efficiency. To address these concerns, this study introduces a personalized dispatching framework that integrates multimodal understanding with pre-training-based fusion techniques. Utilizing Meta’s Llama 2 as the pre-training model, we have developed a fusion algorithm that integrates data from diverse modalities—i.e. text, image, video, and audio—with multi-dimensional worker profiles. This pre-training fusion framework enables our system to interpret and process the multi-level demands of work orders, facilitating rapid and precise dispatching tailored to the skills, locations, and availability of workers, as well as the urgency and type of work orders. Experimental evaluation on a large-scale property service dataset indicates our method’s superiority over traditional rule-based systems, achieving an approximate 19% increase in dispatch accuracy and 20% increase in dispatch speed. Moreover, our model optimizes workers’ routes, supporting the processing of multiple work orders and routes, thus enhancing dispatching efficiency. This study not only offers an innovative approach to personalized dispatching of property service orders but also illustrates the practical potential of pre-training-based fusion techniques. Future research will divide into optimizing fusion strategies and expanding this approach to a wider range of service dispatching scenarios.
Traditional machine learning techniques and dynamic and static analytics are unable to reliably identify frequently changing XSS payloads. This work uses the pre-trained BERT model in Natural Language Processing (NLP) technology for initial feature extraction, and its output is segmented. It then offers a feature splicing approach using fused models to efficiently distinguish between legitimate and malicious XSS payloads. We then use Deep Learning (DL) techniques to further improve the features, and we concatenate the results to detect XSS assaults. This paper uses real data sets from many reliable websites, including owasp, portswigger, and xssed, for experiments with 120,000 normal samples and 15,000 malicious samples to demonstrate the efficacy of our strategy. The experimental results demonstrate the effectiveness of the model fusion method in leveraging the benefits of numerous models to enhance the stability and accuracy of the detection in real scenarios, reaching recall, accuracy, and precision, as well as F1 scores above 99.5%. The method suggested in this paper can identify XSS assaults more successfully than the existing methods discussed in the literature.
Risk management is a crucial link in agricultural product supply chain management. By utilizing digital intelligence technology to analyze and grasp the laws of risk development in Guangdong lychee supply chain, we can effectively prevent and avoid risks, and promote the stable development of the supply chain network. Based on this, this study first identified the risk factors of the lychee supply chain in Guangdong Province through expert interviews and literature review; Secondly, factor analysis, analytic hierarchy process, and fuzzy comprehensive evaluation methods are comprehensively used to evaluate the risks of the supply chain; Finally, based on the results of risk assessment and combined with digital intelligence technology, a risk response strategy is proposed.ct.
There are some problems in the service quality of H college canteens. According to the feedback of students, it is difficult to meet the needs of students. Therefore, the research on service quality evaluation of canteens in H University is helpful to the development of canteens.On the basis of literature research and student feedback system, this paper extracts evaluation indicators with the help of LDA (Latent Dirichlet Allocation) text mining method, designs questionnaires according to preliminary evaluation indicators, collects questionnaire data, and determines the evaluation index system of college canteen service quality through the reliability and validity analysis of the questionnaires. It lays a foundation for the evaluation of canteens.
This study focuses on the relationship between organizations in emergency management of public health emergencies, builds the network of emergency management in emergency plan and in actual emergency response actions, and then uses Social Network Analysis to quantitatively analyses the structure of the two networks from the three levels of overall structure, substructure and individual structure, and verifies whether the network defined by the emergency plan is achieved based on the comparison of the two network structures, so as to explore the problems existing in the network structure and inter-organizational relationships, and put forward optimized countermeasures to improve emergency efficiency.
With the increasing digitization and modernization of enterprise management, employee fingerprint identification management systems have become a key tool to ensure the safety and efficiency of the office space. Such a system not only accurately verifies the identity of employees but can also be used for attendance, access control and other related applications. However, improving the accuracy of fingerprint recognition remains a challenge. In this paper, we propose a fingerprint identification method based on a bidirectional ResNext network and combined with triplet loss. Compared with the traditional ResNext network, our method captures richer fingerprint features through bidirectional structure and triplet loss, thereby achieving higher matching accuracy. We conducted extensive experiments on the public FVC (Fingerprint Verification Competition) dataset and our own collected dataset CSFI (Company staff fingerprint identification). Experimental results show that compared with the original ResNext network, our method achieves approximately 3% and 5% performance improvements on the two data sets respectively. These results verify the effectiveness of our proposed method on fingerprint identification tasks and provide strong technical support for the further development of employee fingerprint identification management systems.
In the context of digital economy, the C2M model, as a new e-commerce model to meet customers’ personalized needs, provides more efficient, convenient and personalized services. In order to explore the evaluation of C2M e-commerce service quality, a C2M e-commerce service quality evaluation model based on LDA theme model and BiLSTM model is proposed. Deep learning has powerful feature selection and feature extraction capabilities, which can automatically capture more abstract and comprehensive semantic information of text, combine topic mining with emotion analysis, and help to identify the needs of users. The LDA theme model is used to extract 8 C2M e-commerce service quality themes; analyzing the service quality themes extracted by LDA based on BiLSTM model, the accuracy, recall rate and F1 factor of Bi-LSTM model are above 90%, indicating that it can accurately predict the emotional tendency of the text, and further select the themes with more negative comments. According to the emotion analysis results, consumers have high negative emotional tendency of commodity cost performance, quality control, personalized customization and logistics service, which are 0.28,0.26,0.25 and 0.21 respectively. Analyanalyze the causes of negative emotions, and put forward targeted suggestions to improve the service quality of C2M e-commerce. The method of this paper makes the C2M e-commerce comment text more expressive, which helps the C2M e-commerce company to better understand the needs of customers and improve the service quality.
This paper introduces an analytical framework of modeling, analysis and improvement of the 5G ambulance-based Emergency Medical Service (EMS) system with queuing network model. The system performance indicators of ambulances and walk-in patients are evaluated and analyzed, especially the ambulance offload delay. Both the average ambulance offload delay in a single hospital and the average total ambulance offload delay in the whole system decrease monotonically with the 5G ambulance penetration rate, which refers to the percentage of ambulances equipped with 5G technology. The applicability of the model is demonstrated and the impact of the 5G ambulance penetration rate is verified through numerical experiments based on real data. This paper analyzes the impact of the 5G ambulance penetration rate on the Emergency Medical Service system, and provides a quantitative analysis method for hospital management and operation.
Because the upstream suppliers of polyester yarn (POY) belong to oligopoly enterprises, the bargaining power of middle and downstream enterprises is low, coupled with the impact of the epidemic and the turmoil of the international situation, the price of POY is constantly fluctuating, which has a greater impact on the procurement cost of warp knitting manufacturers. Therefore, it is of great practical significance to predict the price change of POY, forecast the demand of enterprises and formulate the optimal procurement plan, which can effectively reduce the total cost of enterprises and enable enterprises to develop steadily and healthily. In this paper, a time series model is adopted, taking POY prices from 2016 to 2020 as samples, and the price of POY in 2021 is forecasted and analyzed, while AUTOARIMA is used to determine the optimal ARIMA model parameters. The POY price trend in 2021 is predicted according to the model, and then the production demand of the enterprise is predicted. With the lowest comprehensive procurement cost as the decision-making purpose, the procurement time node arrangement and procurement quantity allocation are completed respectively, and the corresponding strategic procurement plan is formulated to provide reference for the enterprise to formulate the optimal material procurement strategy.
