教学课程资源库 更多>>

关键词： Knowledge representation   Ontology   Deep learning   Reasoning   Occupational health and safety   KNOWLEDGE   SAFETY  

摘要： Reports of fatal and nonfatal workplace injuries depict severe circumstances involving health and safety in industries. This leads to workers staying away from work in U.S. for an average of 12 days in 2020, implicating in managerial, financial, and organizational losses. In this context, Vision-Based Deep Learning (VBDL) and Knowledge Representation and Reasoning (KRR) allow real-time data retrieval of situations along with the semantic modeling and expressivity of the real world to mitigate injuries. This article presents a framework that interoperates vision-based deep learning and ontology reasoning to identify adverse working situations, introducing a novel ontology composed of a holistic perspective of workers' health and safety. Moreover, the article provides multi-agent framework modeling to orchestrate the components' interoperability, describing the framework's architecture and deployment in workrooms. As a result, a practical evaluation over five days produced 8395 axioms constituted by 1210 individuals in the ontology, which allowed a temporal analysis of harmful conditions and their multiple overlapping using SPARQL and reasoning rules, particularly relevant to understanding explanations of overexertion, physical and environmental injuries. Therefore, the proposed ontology-based framework corroborates the long-term support in identifying, assessing, and controlling risks in the industries due to a well-defined knowledge model.

关键词： Deep spatio-temporal network (DSTN)   information fusion   interactive attention mechanism   remaining useful life (RUL)   MODEL  

摘要： Emerging multisource data provide a promising way to make breakthroughs in remaining useful life prediction. Due to the diversity in industrial sites and the complexity of the engineering systems, a large amount of degradation information of machinery is hidden in multitype data, which poses a challenge to adequately capture the complex features that jointly affect remaining useful life. To this end, we propose an interactive attention-based deep spatio-temporal network to effectively fuse vibration waveforms and time-varying operating signals. Specifically, the spatio-temporal structure in the proposed model has the ability to mine long-term dependence and local spatial information from raw multisource data simultaneously. An interactive attention mechanism is used to weight the extracted feature contributions from different source dynamically. Furthermore, a modified mean absolute percentage error criterion is designed in the training process for the inherent properties of the remaining useful prediction. For illustration, a case study of a rotating machinery in an oil refinery and a public dataset of an aircraft engine are investigated. The extensive experiments have demonstrated that, compared to relying solely on either vibrational or operating signals and different fusion strategies, the proposed model can effectively integrate multisource data to reduce prediction loss with an acceptable performance.

关键词： Training   Noise measurement   Data models   Computational modeling   Estimation   Predictive models   Computer science   Instance-dependent noisy labels   time-consistent curriculum learning   image classification  

摘要： Many machine learning algorithms are known to be fragile on simple instance-independent noisy labels. However, noisy labels in real-world data are more devastating since they are produced by more complicated mechanisms in an instance-dependent manner. In this paper, we target this practical challenge of Instance-Dependent Noisy Labels by jointly training (1) a model reversely engineering the noise generating mechanism, which produces an instance-dependent mapping between the clean label posterior and the observed noisy label and (2) a robust classifier that produces clean label posteriors. Compared to previous methods, the former model is novel and enables end-to-end learning of the latter directly from noisy labels. An extensive empirical study indicates that the time-consistency of data is critical to the success of training both models and motivates us to develop a curriculum selecting training data based on their dynamics on the two models' outputs over the course of training. We show that the curriculum-selected data provide both clean labels and high-quality input-output pairs for training the two models. Therefore, it leads to promising and robust classification performance even in notably challenging settings of instance-dependent noisy labels where many SoTA methods could easily fail. Extensive experimental comparisons and ablation studies further demonstrate the advantages and significance of the time-consistency curriculum in learning from instance-dependent noisy labels on multiple benchmark datasets.

关键词： Black -owned banks   Mortgage credit   Residential racial segregation   Subprime lending boom   UNITED-STATES   SEGREGATION   RETHINKING   SOCIOLOGY   RETURNS   CREDIT   ACCESS   WHITE   DEBT  

摘要： Today, black-owned banks are important financial resources challenging economic exclusion. Nevertheless, they do not associate strongly with building black wealth. Some scholars argue this signals black-owned banks are ornamental, or ineffective responses to legacies of economic exclusion in black segregated neighborhoods. To engage these critiques, I draw on the dialectical theoretical frames of cultural assets and structural deficits to examine the effectiveness of blackowned banks during the subprime lending boom-a period when bank practices exploiting a history of economic exclusion in black segregated neighborhoods intensify. Specifically, I analyze administrative data from the Federal Deposit Insurance Corporation (FDIC) and the Home Mortgage Disclosure Act (HMDA) to assess whether black-owned banks associate with access to mortgage credit when the subprime lending boom peaks in 2006. Using propensity score matching with inverse probability weighting, I find black-owned banks do not associate with mortgage originations in 2006; but neighborhoods with black-owned banks receive fewer subprime mortgage loans, compared to matched ones without them. As such, black-owned banks appear to effectively shield black segregated neighborhoods from the time period's predation. Overall, findings imply black-owned banks support protective credit markets during periods of intensifying economic exclusion and exploitation.

关键词： Members Of Society   Science Program   Technology Program   Programs Worldwide   Information System   University Degree   Rapid Changes   Quality Assurance   Computer Science   Technological Change   Computer Technology   Software Engineering   Information Security   Engineering Technology   Community Building   Professional Societies   Student Learning Outcomes   Development Of Artificial Intelligence   Disciplinary Expertise   Program Criteria   Joint Task Force  

摘要： ABET and its member society CSAB are looking into the future, and their direction aims to be responsive to changing needs.

关键词： building management systems   systematic literature review   buildings   smart buildings   computer science   LOAD MANAGEMENT   ENERGY   AUTOMATION   IOT   ARCHITECTURE   BIM   IMPLEMENTATION   MIDDLEWARE   FRAMEWORK   DESIGN  

摘要： Buildings are complex structures composed of heterogeneous elements; these require building management systems (BMSs) to dynamically adapt them to occupants' needs and leverage building resources. The fast growth of information and communication technologies (ICTs) has transformed the BMS field into a multidisciplinary one. Consequently, this has caused several research papers on data-driven solutions to require examination and classification. This paper provides a broad overview of BMS by conducting a systematic literature review (SLR) summarizing current trends in this field. Unlike similar reviews, this SLR provides a rigorous methodology to review current research from a computer science perspective. Therefore, our goal is four-fold: (i) Identify the main topics in the field of building; (ii) Identify the recent data-driven methods; (iii) Understand the BMS's underlying computing architecture (iv) Understand the features of BMS that contribute to the smartization of buildings. The result synthesizes our findings and provides research directions for further research.

关键词： Computer Science Jobs   Skillsets Required   Hard and Soft Skills   Social Network Analysis   Community analysis  

摘要： The rapid growth of technology and computer science, which has led to a surge in demand for skilled professionals in this field. The skill set required for computer science jobs has evolved rapidly, creating challenges for those already in the workforce who need to adapt their skills quickly to meet industry demands. To stay ahead of the curve, it is essential to understand the hottest skills needed in the field. The article introduces a new method for analyzing job advertisements using social network analysis to identify the most critical skills required by employers in the *** this research, to form the communication network of skills, first 5763 skills were collected from the LinkedIn social network, then the relationship between skills was collected and searched in 7777 computer science job advertisements, and finally, the balanced communication network of skills was formed. The study analyzes the formed communication network of skills in the computer science job market and identifies four distinct communities of skills: Generalists, Infrastructure and Security, Software Development, and Embedded Systems. The findings reveal that employers value both hard and soft skills, such as programming languages and teamwork. Communication skills were found to be the most important skill in the labor market. Additionally, certain skills were highlighted based on their centrality indices, including communication, English, SQL, Git, and business skills, among others. The study provides valuable insights into the current state of the computer science job market and can help guide individuals and organizations in making informed decisions about skills acquisition and hiring practices.

关键词： Codes   Linear codes   Codecs   Cryptography   Telecommunications   Computer science   Hamming weight   Quaternary code   simplicial complex   gray map   binary nonlinear code   minimal code   secret sharing scheme   LINEAR CODES  

摘要： Recently, simplicial complexes are used in constructions of several infinite families of minimal and optimal linear codes by Hyun et al. Building upon their research, in this paper more linear codes over the ring Z(4) are constructed by simplicial complexes. Specifically, the Lee weight distributions of the resulting quaternary codes are determined and two infinite families of four-Lee-weight quaternary codes are obtained. Compared to the databases of Z(4 )codes by Aydin et al., at least nine new quaternary codes are found. Thanks to the special structure of the defining sets, we have the ability to determine whether the Gray images of certain obtained quaternary codes are linear or not. This allows us to obtain two infinite families of binary nonlinear codes and one infinite family of binary minimal linear codes. Furthermore, utilizing these minimal binary codes, some secret sharing schemes as a byproduct also are established.

关键词： Life satisfaction   Being accepted by others   Community identity   Relocated residents for poverty alleviation (RRPA)   Random intercept cross-lagged panel model (RI-CLPM)   SELF-DETERMINATION THEORY   SOCIAL EXCLUSION   MOTIVATION   INCOME   WELL  

摘要： In China, the ultimate goal of the relocation for poverty alleviation (RPA) policy is to ensure that the relocated poor smoothly relocate, stably resettle, and become wealthier. With the completion of smooth relocation, the next crucial task is to guarantee that residents relocated for poverty alleviation (RRPA) maintain stability. Given that RRPA's life satisfaction affects their willingness and behavior to stabilize their placement. This study aims to explore whether being accepted by others and community identity are related to RRPA life satisfaction, and whether these three variables have a mutual impact over time. Based on the social identity and need to belong theories, a model of the interactive influence of being acceptance accepted by others, community identity, and life satisfaction was constructed. Using a random intercept cross-lagged panel model (RI-CLPM), 383 RRPA datasets were analyzed. The results showed that being accepted by others significantly predicted life satisfaction for RRPA. Being accepted by others and community identity, as well as community identity and life satisfaction, can predict each other on cross-lagged paths. This suggests that resorting to community identity as an intermediate variable, being accepted by others and life satisfaction may form a virtuous cycle over time.

关键词： Learning environments   Instructional modality   Motivation   Self-regulated learning   Academic achievement   Blended computer science education   SELF-EFFICACY   STUDENTS MOTIVATION   TIME MANAGEMENT   PROCRASTINATION   PERFORMANCE   STRATEGIES   ONLINE  

摘要： The goal of the present study was to explore the relations among perceived psychosocial learning environments, instructional modality, motivation, self-regulated learning, and academic achievement in blended computer science education. The participants were 207 undergraduate students enrolled in a blended online and face-to-face design course. We employed exploratory structural equation modeling to analyze the data. Our findings indicated that within the perceived psychosocial learning environments, dimensions such as personal relevance, authentic learning, and active learning significantly predicted various motivational beliefs. Furthermore, active learning and expectancy were strong predictors of effective time management, while instructional modality and emotional cost were closely linked to academic procrastination. Notably, academic procrastination emerged as the sole significant predictor of academic achievement, measured by course grades. These results suggest that traditional curricula emphasizing textbook reading and code memorization may be ineffective in teaching computer science. Additionally, our study highlights a higher tendency for procrastination in online settings. We recommend a curriculum focused on personal relevance, authentic learning, and active learning to better motivate students and enhance their ability to manage their learning effectively. To improve academic achievement in computer science education, it is crucial to address maladaptive self-regulatory processes and motivational beliefs, which primarily arise from active learning and instructional modality. We will conclude with specific recommendations for designing learning environments that better support computer science education.