Crowdsourcing is a critical technology in social manufacturing, which leverages an extensive and boundless reservoir of human resources to handle a wide array of complex tasks. The successful execution of these complex tasks relies on task decomposition (TD) and allocation, with the former being a prerequisite for the latter. Recently, pretrained language models-based methods have garnered significant attention. However, they are constrained to handling straightforward common-sense tasks due to their inherent restrictions involving limited and difficult-to-update knowledge as well as the presence of “hallucinations.” To address these issues, we propose a retrieval-augmented generation-based crowdsourcing framework that reformulates TD as event detection from the perspective of natural language understanding. However, the existing detection methods fail to distinguish differences between event types and always depend on heuristic rules and external semantic analyzing tools. Therefore, we present a prompt-based contrastive learning framework for TD (PBCT), which incorporates a prompt-based trigger detector to overcome dependence. Additionally, trigger-attentive sentinel and masked contrastive learning are designed to provide varying attention to trigger and contextual features according to different event types. Extensive experiment results demonstrate our method is highly competitive in both supervised and zero-shot detection. A case study on printed circuit board design and manufacturing is used to validate its adaptability and scalability in unfamiliar professional domains.

In traditional binary classification tasks, learning algorithms conventionally distinguish positive and negative samples by leveraging fully labeled training data. However, in practical applications, there frequently arises a scenario where only a small number of positive samples and a large volume of unlabeled data exist, with the latter potentially containing a mix of positive and negative instances. This situation is known as positive-unlabeled learning (PUL) and has drawn considerable interest across various domains, such as text categorization. Despite numerous studies addressing PUL issues, few have focused on improving the two-step methodology in the data partitioning process, and even fewer have explored the application of genetic programming (GP) to PUL challenges. This article introduces a GP-based embedded unlabeled data partitioning method (EPGP) tailored for the PUL problem, particularly in the context of few-shot learning scenarios. The approach adopts a multiphase data partitioning strategy, integrating the partitioning process into the evolutionary cycle of the GP population, progressively isolating positive samples from the unlabeled data to yield a PU dataset closer to the real-world distribution. To achieve smoother data partitioning, a dynamically adjusted partitioning threshold strategy is incorporated. Finally, an ensemble method is devised, capitalizing on the high confidence associated with the originally labeled positive samples to generate the final classification outcome via weighted voting. Experimental evaluations of EPGP against other state-of-the-art PUL methods on 22 datasets show significant advantages in terms of balanced accuracy and macro-F1 scores on 17 datasets. Moreover, a case study on text data classification tasks demonstrates that EPGP consistently delivers substantial performance enhancements under varying proportions of labeled positive samples.

Aiming to boost production efficiency and reduce human workload, human-centricity has emerged as the core concept of Industry 5.0 (I5.0). However, current works have not established a unified automation and autonomous framework for human-centric smart manufacturing across various real world applications. Addressing this gap, this research introduces an innovative automated framework, ParallelWorkforce, which integrates blockchain intelligence and decentralized autonomous organizations and operations (DAOs) to drive the evolution from digital twins to parallel intelligence. First, this research conducts a comprehensive investigation into smart manufacturing in I5.0, summarizing the ongoing evolution. Next, a detailed exploration of ParallelWorkforce is provided to offer customized strategies for managing different levels of out-of-distribution events, significantly alleviating the workload on biological workers and maximizing the potential of both digital and robotic workers. Finally, the development of ParallelWorkforce across various key applications of smart manufacturing is demonstrated, including autonomous transportation, task assignment, and worker management. This research provides a viable solution for the further development of human-centered smart manufacturing and paves the way for the realization of “6S” goals in I5.0.

In recent years, more and more scholars have begun to focus on sentiment analysis on social media. Current sentiment analysis collects all relevant data, including public thoughts, opinions, and feelings, from a variety of open sources. In addition, it automatically predicts different aspects of outcomes or trends based on information collected globally in real time. This research area explores how to extract sentiment information from different modalities (e.g., text, images, and audio). However, the currently existing techniques face several challenges. It is difficult to achieve effective interaction with completely heterogeneous data, and these techniques cannot adequately guarantee data security during data interaction, which is particularly important when dealing with sensitive information. Therefore, this article introduces existing methods for protecting data privacy. Based on this foundation, we propose a novel algorithm called collaborative dynamic hierarchical federated learning with hypernetwork aggregation (CDHFL-HA), which is suitable for sentimental analysis. CDHFL-HA ensures that the data remain local to each participant while leveraging the data similarity between participants on the server and processing interference data on the participant to enhance the accuracy of the current sentimental analysis. In addition, an essential aspect considered in the proposed algorithm is explainability. Understanding the decisions and predictions made by sentiment analysis models is crucial for gaining trust and acceptance in real-world applications. CDHFL-HA incorporates explainability features, providing insights into the decision-making process, thus enhancing the interpretability of sentiment analysis results. Numerous experimental results show that the algorithm outperforms existing algorithms in complex scenarios, with a minimum accuracy of 0.6007 and a maximum of 0.9962. In addition, it can be seen from the experimental results in this article, that the communication parameters in the experiments are similar to those of other federated learning, while the number of training rounds is improved by up to 50% (i.e., 20 rounds faster) relative to other algorithms.

Many people have been severely affected by the COVID-19 pandemic, which has caused intense anxiety, fear, and complex feelings or emotions. People’s emotions have changed and become more complicated since coronavirus vaccinations were introduced. Sentiment analysis of COVID-19 vaccination is critical for understanding public perception, vaccine hesitancy, monitoring vaccine impact, identifying adverse reactions, making policies, and allocating resources. Some artificial intelligence (AI)-based systems have been reported in the literature to analyze the sentiment of COVID-19 vaccination. However, most of them are end-to-end models that require explanation for their results in identifying COVID-19 vaccination sentiment. An explainable AI-based model can improve decision-making, transparency, and interpretability. It can enable users to comprehend how the model makes predictions and the factors that influence the outcome. Therefore, this study suggests a COVID-Twitter-BERT and LSTM (CT-BERT-LSTM) staking that is explicable for determining people’s views on the COVID-19 vaccination. The prediction of the proposed CT-BERT-LSTM model is then examined to determine where the suggested system successfully learned the context of the tweet and where it failed to do so. The proposed CT-BERT-LSTM model performed exceptionally well and outperformed the existing models with a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -score of 0.88 in the sentiment identification of the COVID-19 vaccination.

This study explores an innovative approach to predicting individual life satisfaction by combining large language models (LLMs) with machine learning (ML) techniques. Traditional life satisfaction assessments rely on self-report questionnaires, which can be time-consuming and resource intensive. To address these limitations, we developed a method that utilizes LLMs for feature extraction from open-ended self-statement texts, followed by ML prediction. We compared this approach with standalone LLM predictions and expert ratings. A sample of 378 participants completed the satisfaction with life scale (SWLS) and wrote self-statements about their current life situation. The LLM-based ML model, using a LightGBM regressor, achieved a correlation of 0.542 with self-reported SWLS scores, outperforming both the standalone LLM (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</i> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$=$</tex-math></inline-formula> 0.491) and expert ratings (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</i> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$=$</tex-math></inline-formula> 0.455). Effect size analysis revealed a statistically significant moderate effect size difference between the LLM-based ML model and expert ratings (Cohen's <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</i> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$=$</tex-math></inline-formula> 0.499, 95% CI [0.043, 0.955]). These findings demonstrate the potential of integrating LLM and ML for an efficient and accurate assessment of life satisfaction, challenging conventional methods, and opening new avenues for psychological measurement. The study's implications extend to research, clinical practice, and policymaking, offering promising advancements in AI-assisted psychological assessment.

Online advertising has been the mainstream monetization approach for internet-based companies, in which bid optimization plays a crucial role in enhancing advertising performance. Currently, the bid optimization problem has narrowed down to two specific forms: Budget-constrained bidding (BCB) and Multi-constraint bidding (MCB). Existing solutions try to solve BCB/MCB via linear programming solvers, learning methods, or feedback control. However, in large-scale complex e-commerce, they still suffer from inefficiency, poor convergence, or slow adaptation to the changing market. This research presents an online receding optimization method as a solution for practical bid optimization problems. We conduct a theoretical analysis of the optimal bidding strategy's structure. Further, an online receding optimization process is designed based on open-loop feedback control, which periodically updates a constructed optimal bid formulation that can be solved by linear programming. Then, considering large-scale linear programming problems, we propose an efficient down sampling scheme. Besides, a neural-network-based auction scale prediction is used to adapt to the changing market. Finally, a series of online A/B experiments on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Taobao Sponsored Search</i> compare our work to industrial methods and state-of-the-art from several aspects. The proposed method has been implemented on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Taobao</i>, a billion-scaled online advertising business, for over a year.

Accurately predicting the future trajectories of traffic agents is paramount for autonomous unmanned systems, such as self-driving cars and mobile robotics. Extracting abundant temporal and social features from trajectory data and integrating the resulting features effectively pose great challenges for predictive models. To address these issues, this article proposes a novel multibranch attentive transformer (MBAT) trajectory prediction network for traffic agents. Specifically, to explore and reveal diverse correlations of agents, we propose a decoupled temporal and spatial feature learning module with multibranch to extract temporal, spatial, as well as spatiotemporal features. Such design ensures each branch can be specifically tailored for different types of correlations, thus enhancing the flexibility and representation ability of features. Besides, we put forward an attentive transformer architecture that simultaneously models the complex correlations possibly occurring in historical and future timesteps. Moreover, the temporal, spatial, and spatiotemporal features can be effectively integrated based on different types of attention mechanisms. Empirical results demonstrate that our model achieves outstanding performance on public ETH, UCY, SDD, and INTERACTION datasets. Detailed ablation studies are conducted to verify the effectiveness of the model components.