Journal of Organizational and End User Computing

With the growing urgency of global climate change, carbon neutrality, as a strategy to reduce greenhouse gas emissions into the atmosphere, is increasingly seen as a critical solution. However, current forecasting models still face significant challenges and limitations in accurately and effectively predicting carbon emissions and their associated effects. These challenges largely stem from the complexity of carbon emission data and the interplay of anthropogenic and natural factors. To overcome these obstacles, the authors introduce an advanced forecasting model, the SSA-Attention-BIGRU network. This model ingeniously integrates an external attention mechanism, bidirectional GRU, and SSA components, allowing it to synthesize various key factors and enhance prediction accuracy when forecasting carbon neutrality trends. Through experiments on multiple datasets, the results demonstrate that, compared to other popular methods, the SSA-Attention-BIGRU network significantly excels in prediction accuracy, robustness, and reliability.

Innovation management involves planning, organizing, and controlling innovation within an organization, while venture capital evaluation assesses investment opportunities in startups and early-stage companies. Both fields require effective decision-making and data analysis. This study aims to enhance innovation management and venture capital evaluation by combining CNN and GRU using deep learning. The approach consists of two steps. First, the authors build a deep learning model that fuses CNN and GRU to analyze diverse data sources like text, finance, market trends, and social media sentiment. Second, they optimize the model using the gorilla troop optimization (GTO) algorithm, inspired by gorilla behavior. GTO efficiently explores the solution space to find optimal or near-optimal solutions. The authors compare the fused CNN-GRU model with traditional methods and evaluate the GTO algorithm's performance. The results demonstrate improvements in innovation management and venture capital evaluation.

With the rapid development of artificial intelligence and deep learning, image-text matching has gradually become an important research topic in cross-modal fields. Achieving correct image-text matching requires a strong understanding of the correspondence between visual and textual information. In recent years, deep learning-based image-text matching methods have achieved significant success. However, image-text matching requires a deep understanding of intra-modal information and the exploration of fine-grained alignment between image regions and textual words. How to integrate these two aspects into a single model remains a challenge. Additionally, reducing the internal complexity of the model and effectively constructing and utilizing prior knowledge are also areas worth exploring, therefore addressing the issues of excessive computational complexity in existing fine-grained matching methods and the lack of multi-perspective matching.

This article presents a study using ResNet-50, GRU, and transfer learning to construct a marketing decision-making model and predict consumer behavior. Deep learning algorithms address the scale and complexity of consumer data in the information age. Traditional methods may not capture patterns effectively, while deep learning excels at extracting features from large datasets. The research aims to leverage deep learning to build a marketing decision-making model and predict consumer behavior. ResNet-50 analyzes consumer data, extracting visual features for marketing decisions. GRU model temporal dynamics, capturing elements like purchase sequences. Transfer learning improves performance with limited data by using pre-trained models. By comparing the model predictions with ground truth data, the performance of the models can be assessed and their effectiveness in capturing consumer behavior and making accurate predictions can be measured. This research contributes to marketing decision-making. Deep learning helps understand consumer behavior, formulate personalized strategies, and improve promotion and sales. It introduces new approaches to academic marketing research, fostering collaboration between academia and industry.

In this study, the authors explore how financial institutions make decisions about stock trading strategies in a rapidly changing and complex environment. These decisions are made with limited, often inconsistent information and depend on the current and future strategies of both the institution itself and its competitors. They develop a dynamic game model that factors in this imperfect information and the evolving nature of decision-making. To model reward transitions, they utilize a combination of t-Copula simulation of a non-stationary Markov chain, probabilistic fuzzy regression, and chaos optimization algorithms. They then apply deep q-network, a method from deep reinforcement learning, to ensure the effectiveness of the chosen strategy during ongoing decision-making. The approach is significant for both researchers across fields and practical professionals in the finance industry.

When compared to traditional indicators, text information can capture market sentiment, investor confidence, and public opinion more effectively. Meanwhile, the mixed-frequency dynamic factor model (MF-DFM) can capture current changes. In this study, the authors constructed a financial cycle measurement and nowcasting framework by incorporating text information into factors derived from MF-DFM. The findings reveal that, first, the financial cycle indicator (FCI) provides a more detailed and forward-looking perspective on major events. Second, it can serve as an effective “early warning system” by cross-referencing economic indicators. Third, financial cycles exhibit five short cycles, with contraction periods being longer than expansion phases and expansion amplitudes surpassing contractions. Lastly, the analysis suggests a potential turning point in the second half of 2023. This research represents a valuable attempt to integrate big data for more sensitive, timely, and accurate monitoring of financial dynamics.

The results indicate a dynamic pattern of interconnectedness throughout history. Based on the findings, the transmission of volatility exhibited a higher magnitude during the period of COVID-19. The issue of high transmission volatility due to limited diversification options concerns investors, green stakeholders, and policymakers alike. This article proposes various potential areas for future research. The ICEA index can potentially assist businesses operating in environmentally sensitive sectors make well-informed policy decisions. It includes sectors such as environmental green bonds, and commodities. Consideration should be given to implementing blockchain technology, as it can consume less power in this particular scenario. By employing a time-frequency paradigm, this study is able to incorporate the investment horizon, a crucial factor to be taken into account when making financial judgments. The advancement of this research could be facilitated by directing our attention toward the implications of our findings on portfolios and developing appropriate measures for their evaluation.

This study aims to enhance the efficacy of personalized learning paths by amalgamating transformer models, generative adversarial networks (GANs), and reinforcement learning techniques. To refine personalized learning trajectories, the authors integrated the transformer model for enhanced information assimilation and learning path planning. Through generative adversarial networks, the authors simulated the fusion and interaction of multi-modal information, refining the training of virtual teaching assistants. Lastly, reinforcement learning was employed to optimize the interaction strategies of these assistants, aligning them better with student needs. In the experimental phase, the authors benchmarked their approach against six state-of-the-art models to assess its effectiveness. The experimental outcomes highlight significant enhancements achieved by the authors’ virtual teaching assistant compared to traditional methods. Precision improved to 95% and recall to 96%, and an F1 score exceeding 95% was attained.

This project addresses demand forecasting and inventory optimization in supply chain management. Traditional methods have limitations with complex demand patterns and large-scale data. Deep learning techniques are employed to enhance accuracy and efficiency. The project utilizes BO-CNN-LSTM, leveraging Bayesian optimization for hyperparameter tuning, Convolutional Neural Networks (CNNs) for spatiotemporal feature extraction, and Long Short-Term Memory Networks (LSTMs) for modeling sequential data. Experimental results validate the effectiveness of the approach, outperforming traditional methods. Practical implementation in supply chain management improves operational efficiency and cost control.

Effectively fusing information between the visual and language modalities remains a significant challenge. To achieve deep integration of natural language and visual information, this research introduces a multimodal fusion neural network model, which combines visual information (RGB images and depth maps) with language information (natural language navigation instructions). Firstly, the authors used faster R-CNN and ResNet50 to extract image features and attention mechanism to further extract effective information. Secondly, GRU model is used to extract language features. Finally, another GRU model is used to fuse the visual- language features, and then the history information is retained to give the next action instruction to the robot. Experimental results demonstrate that the proposed method effectively addresses the localization and decision-making challenges for robotic vacuum cleaners.

As global financial markets continue to evolve and change, financial risk monitoring and early warning have become increasingly important. However, the complexity and diversity of financial markets have led to the emergence of multidimensional and multimodal data. Traditional risk monitoring methods face difficulties in handling such diverse data and adapting to the monitoring and early warning needs of emerging risk types. To address these issues, this article proposes a financial risk intelligent monitoring and early warning model that integrates deep learning to better cope with uncertainty and risk in the financial market. Firstly, the authors introduce an LSTM model in the initial approach, trained on historical financial market data, to capture long-term dependencies and trends in the data, enabling effective monitoring of financial risk. They also optimize the model architecture to improve its performance and prediction accuracy. Secondly, the authors further introduce a transformer model with self-attention mechanism to better handle sequential data.

The fairness of vocational contest scoring is key to generating reliable competency assessments. This study examined the performance impact of the motivation of English-as-a-foreign-language learners in contests with vocabulary knowledge antecedents in the contexts of artificial intelligence (AI) and blockchain (BC). The sample comprised 185 participants of an oral English contest at higher vocational institution in China. AI-powered scoring of learners' contest performance and a survey were used to collect data. The findings revealed that learners’ intrinsic drive was the main positive factor, outweighing their extrinsic motivation, and that AI and BC increased the trustworthiness and integrity of contest records, thus providing new opportunities to build learner trust and form psychological incentives. This study enriches foreign language motivation theory in the context of contest research and highlights the importance of using AI and BC to enhance the scoring accuracy and credibility of contests as authoritative evaluation instruments in vocational education.

In the face of intensifying global climate change, carbon neutrality has emerged as a pivotal strategy to curb greenhouse gas emissions and confront the complexities associated with climate challenges. However, achieving carbon neutrality poses a formidable challenge: the identification and mitigation of anomalies within the carbon sequestration process. These anomalies can result in unintended carbon dioxide leakage, emissions, or system failures, thus jeopardizing the feasibility and resilience of carbon neutrality initiatives. This research introduces the ResNet-BIGRU-TPA network, an innovative model that integrates deep learning techniques with time series attention mechanisms. The primary focus centers on addressing the intricate task of anomaly detection within the realm of carbon offsetting, specifically aiming to enhance precision in identifying a wide array of complex anomalous events. Through rigorous experimental validation across four diverse datasets, the model has exhibited exceptional performance.

Visual search technology, because of its convenience and high efficiency, is widely used by major tourism e-commerce platforms in product search functions. This study introduces an innovative visual search engine model, namely CLIP-ItP, aiming to thoroughly explore the application potential of visual search in tourism e-commerce. The model is an extension of the CLIP (contrastive language-image pre-training) framework and is developed through three pivotal stages. Firstly, by training an image feature extractor and a linear model, the visual search engine labels images, establishing an experimental visual search engine. Secondly, CLIP-ItP jointly trains multiple text and image encoders, facilitating the integration of multimodal data, including product image labels, categories, names, and attributes. Finally, leveraging user-uploaded images and jointly selected product attributes, CLIP-ItP provides personalized top-k product recommendations.

Conventional automobile manufacturing plants involve intricate assembly, testing, and debugging processes heavily reliant on manual operations. This study aims to explore the application of industrial internet of things (IIoT) and deep learning algorithms to achieve process automation in manufacturing. Firstly, utilizing IIoT technology, OPC UA, and point cloud fitting techniques, a comprehensive modeling of most equipment and materials within the factory is conducted, constructing a digital twin (DT) model as a virtual representation of actual equipment. Subsequently, the study innovatively introduces the deep Q network algorithm, facilitating the automatic transition of the production process and improving production efficiency. Through comparison with ten baseline models, the proposed model demonstrates an improvement in production efficiency of at least four percentage points compared to other models. Experimental validation confirms the effectiveness of the proposed model in the smart factory for electric vehicle manufacturing.

The existing big data analytics measures were developed without considering the cultural dimensions of developing countries. This research aims to develop and validate measures for big data Vs and cultural big data analytics and study their impacts on the developing countries’ big data value proposition. Following MacKenzie’s and Shiau and Huang’s scale development procedures, data was collected twice from individuals in a developing country to refine the scale and reexamine its properties. PLS methods were used to study the impacts of big data Vs and cultural big data analytics on the value proposition. The findings revealed that big data analytics snobbism and conformism positively impact big data value proposition. Similarly, big data volume, velocity, and variety positively impact the value proposition. Paradoxically, big data veracity and variability do not significantly affect the value proposition. Surprisingly, big data analytics fatalism negatively impacts the value proposition. Theoretical and practical contributions were offered.

With the continuous evolution of digital marketing, the generation of advertising images has become crucial in capturing user interest and enhancing advertising effectiveness. However, existing methods face limitations in meeting the diverse and creative demands of advertising content, necessitating innovative algorithms to improve advertising generation outcomes. In addressing these challenges, this study proposes a deep learning algorithm framework that cleverly integrates a generative adversarial network and an VGG-based visual transformer model to enhance the effectiveness of advertising image generation. Systematic experimentation shows that the model proposed in this article achieves an AUC metric value of more than 0.7 on several datasets. The results of the experiments demonstrate that the novel algorithm significantly improves the attractiveness of advertising content, particularly showcasing substantial benefits in website operations during online evaluation experiments.

In today's digital age, the e-commerce industry continues to grow and flourish. The widespread application of computer vision technology has brought revolutionary changes to e-commerce platforms. Extracting image features from e-commerce platforms using deep learning techniques is of paramount importance for predicting product sales. Deep learning-based computer vision models can automatically learn image features without the need for manual feature extractors. By employing deep learning techniques, key features such as color, shape, and texture can be effectively extracted from product images, providing more representative and diverse data for sales prediction models. This study proposes the use of ResNet-101 as an image feature extractor, enabling the automatic learning of rich visual features to provide high-quality image representations for subsequent analysis. Furthermore, a bidirectional attention mechanism is introduced to dynamically capture correlations between different modalities, facilitating the fusion of multimodal features.

With the expanding reach of the Internet of Things, information security threats are increasing, including from the very professionals tasked with defending against these threats. This study identified factors impacting information security behavior among these individuals. Protection motivation theory and the theory of planned behavior were employed along with work-related organizational factors as a theoretical framework. Data were collected through a survey of 595 information security professionals working in Saudi information technology companies. Structural equational modeling was used to analyze the data. Threat susceptibility, threat severity, self-efficacy, response cost, fear attitude, behavioral control, subjective norms, and organizational commitment were found to play a significant role in information security protection motivation and behavior, while job satisfaction did not.

In pursuit of promoting sales, certain e-commerce vendors post scarce inventory messages on product pages to signal impending stockouts. To explore how disclosure information strategy regarding scarce inventory influences online sales, this study utilizes Chinese e-commerce data collected from February 1st to April 30th, 2023 and constructs an empirical model that delves into the relationship between the disclosure of scarce inventory information and online sales based on the signal theory. The empirical findings reveal a positive impact of scarce inventory information disclosure on online sales, with this impact being more pronounced under the moderating role of the commodity discount rate. These results hold substantial theoretical implications and offer valuable insights for practical applications in the e-commerce domain.

To elevate the intelligence of customer service dialogue systems, this article proposes an intelligent customer service system comprising chat dialogue subsystems, task-oriented multi-turn dialogue subsystems, single-turn dialogue subsystems, and an integration model. Firstly, to enhance diversity of responses and improve user experience, particularly in casual chat scenarios, this article presents a Seq2Seq-based approach for multi-answer responses, allowing for more expressive emotional expression in responses. Secondly, to address situations where customers cannot articulate their needs in a single sentence during multi-turn dialogues, this article designs a task-oriented multi-turn dialogue module. It employs intent recognition and slot filling to maintain contextual information throughout the conversation, aiding customers in problem resolution. Lastly, to overcome the current limitation of intelligent customer service models providing relatively one-dimensional answers in specific domains.

Anomaly detection plays a crucial role in the field of machine learning, as it involves constructing detection models capable of identifying abnormal samples that deviate from expected patterns, using unlabeled or normal samples. In recent years, there has been a growing interest in integrating anomaly detection into image processing to tackle challenges related to target detection, particularly when dealing with limited sample availability. This paper introduces a novel fully connected network model enhanced with a memory augmentation mechanism. By harnessing the comprehensive feature capabilities of the fully connected network, this model effectively complements the representation capabilities of convolutional neural networks. Additionally, it incorporates a memory module to retain knowledge of normal patterns, thereby enhancing the performance of existing models for video anomaly detection. Furthermore, we present a video anomaly detection system designed to identify abnormal image data within surveillance videos, leveraging the innovative network architecture described above.

Consumer credit assessment has always been a crucial concern in the financial industry. It involves evaluating an individual's credit history and their ability to repay loans, playing a pivotal role in the risk management and lending decisions made by credit institutions. In the present landscape, traditional credit assessment methods confront various shortcomings. Firstly, they typically only consider static features and are unable to capture the dynamic changes in an individual's credit profile over time. Secondly, traditional methods struggle with processing complex time series data, failing to fully exploit the importance of time-related information. To address these challenges, we propose an innovative solution – the XGBoost-LSTM model optimized with the AdaBound algorithm. This hybrid model combines two powerful machine learning techniques, XGBoost and LSTM, to leverage both static and dynamic features effectively.

This article aims to study the role of digital finance in rural revitalization by utilizing the data for 30 provinces in China Mainland from 2012 to 2019, via the fixed effect model and differential GMM model. The empirical results show that the level of rural revitalization varies among different regions in China. In addition, the development of digital inclusive finance is essential in promoting rural revitalization, which is credible while we conduct several robustness tests such as changing the measurement of digital finance and including the dynamic progress by utilizing the GMM estimation. The impact of digital inclusive finance on rural revitalization is not constant among different regions, the positive impact of digital finance on rural revitalization is stronger in eastern region than that in central and western regions.

With the exacerbation of the rural aging population trend, home-based health monitoring for the rural elderly has become a societal focal point, demanding an effective technological means to elevate the level of rural elderly health management. However, traditional algorithms for monitoring rural elderly behavior face myriad challenges, such as effectively capturing temporal and spatial features. Consequently, addressing the need to enhance the accuracy and robustness of rural elderly behavior recognition has become an urgent problem to solve. This study responds to this challenge by comprehensively employing deep learning and temporal modeling techniques, designing, and validating a short-term and long-term dual-layer home-based health monitoring system for the rural elderly.In the short-term layer, the model utilizes smartphones to collect health information from the rural elderly in various ways and performs real-time anomaly behavior detection.