Search results

Single-shot multi-category clothing recognition and retrieval play a crucial role in online searching and offline settlement scenarios. Existing clothing recognition methods based on RGBD clothing images often suffer from high-dimensional feature representations, leading to compromised performance and efficiency.

To address this issue, this paper proposes a novel method called Manifold Embedded Discriminative Feature Selection (MEDFS) to select global and local features, thereby reducing the dimensionality of the feature representation and improving performance. Specifically, by combining three global features and three local features, a low-dimensional embedding is constructed to capture the correlations between features and categories. The MEDFS method designs an optimization framework utilizing manifold mapping and sparse regularization to achieve feature selection. The optimization objective is solved using an alternating iterative strategy, ensuring convergence.

Empirical studies conducted on a publicly available RGBD clothing image dataset demonstrate that the proposed MEDFS method achieves highly competitive clothing classification performance while maintaining efficiency in clothing recognition and retrieval.

This paper introduces a novel approach for multi-category clothing recognition and retrieval, incorporating the selection of global and local features. The proposed method holds potential for practical applications in real-world clothing scenarios.

Kriging surrogate model has demonstrated a powerful ability to be applied to a variety of engineering challenges by emulating time-consuming simulations. However, when it comes to problems with high-dimensional input variables, it may be difficult to obtain a model with high accuracy and efficiency due to the curse of dimensionality. To meet this challenge, an improved high-dimensional Kriging modeling method based on maximal information coefficient (MIC) is developed in this work.

The hyperparameter domain is first derived and the dataset of hyperparameter and likelihood function is collected by Latin Hypercube Sampling. MIC values are innovatively calculated from the dataset and used as prior knowledge for optimizing hyperparameters. Then, an auxiliary parameter is introduced to establish the relationship between MIC values and hyperparameters. Next, the hyperparameters are obtained by transforming the optimized auxiliary parameter. Finally, to further improve the modeling accuracy, a novel local optimization step is performed to discover more suitable hyperparameters.

The proposed method is then applied to five representative mathematical functions with dimensions ranging from 20 to 100 and an engineering case with 30 design variables.

The results show that the proposed high-dimensional Kriging modeling method can obtain more accurate results than the other three methods, and it has an acceptable modeling efficiency. Moreover, the proposed method is also suitable for high-dimensional problems with limited sample points.

Contextual cues have become a hot research topic in the field of mobile consumer behavior, owing to the continuous rise of digital marketing. However, the complex online shopping scene makes it challenging to directly identify the association between the characteristics of contextual cues and consumer behavior. Presently, few studies have only systematically extracted and refined the types and characteristics of contextual cues. The purpose of this study is to explore the types and mechanisms of contextual cues in online shopping scenarios.

This study uses the word2vec algorithm, grounded theory and co-occurrence cluster method, along with online shopping word-of-mouth (WOM) text and consumer behavior theory, in order to explore different types of contextual cues and its efficiency from 5,619 comment corpus.

This study puts forward the following conclusions. (1) From the perspective of online shopping, contextual cues comprise aesthetic perception cues, value perception cues, trust-dependent cues, time perception cues, memory attention cues, spatial perception cues, attribute cues and relationship cues. (2) Based on the online shopping scenarios, contextual cues and their interaction effects exert an effect on consumer satisfaction, recommendation, purchase and return behavior.

The study conclusions are helpful to further reveal the deep association between contextual cues and consumer behavior in the process of online shopping, thus providing practical and theoretical enlightenment for enterprises to not only effectively reshape the scene but also promote the consumers' active purchase behavior.

It is of a great significance for the health monitoring of a liquid rocket engine to build an accurate and reliable fault prediction model. The thrust of a liquid rocket engine is an important indicator for its health monitoring. By predicting the changing value of the thrust, it can be judged whether the engine will fail at a certain time. However, the thrust is affected by various factors, and it is difficult to establish an accurate mathematical model. Thus, this study uses a mixture non-parametric regression prediction model to establish the model of the thrust for the health monitoring of a liquid rocket engine.

This study analyzes the characteristics of the least squares support vector regression (LS-SVR) machine . LS-SVR is suitable to model on the small samples and high dimensional data, but the performance of LS-SVR is greatly affected by its key parameters. Thus, this study implements the advanced intelligent algorithm, the real double-chain coding target gradient quantum genetic algorithm (DCQGA), to optimize these parameters, and the regression prediction model LSSVRDCQGA is proposed. Then the proposed model is used to model the thrust of a liquid rocket engine.

The simulation results show that: the average relative error (ARE) on the test samples is 0.37% when using LS-SVR, but it is 0.3186% when using LSSVRDCQGA on the same samples.

The proposed model of LSSVRDCQGA in this study is effective to the fault prediction on the small sample and multidimensional data, and has a certain promotion.

The original contribution of this study is to establish a mixture non-parametric regression prediction model of LSSVRDCQGA and properly resolve the problem of the health monitoring of a liquid rocket engine along with modeling the thrust of the engine by using LSSVRDCQGA.

Foundry produces cast metal components and parts for various industries and drives manufacturing excellence all over the world. Assuring quality of these components and parts is vital for the end product quality. The complexity in foundry operations increases with the complexity in designs, patterns and geometry and the quality parameters of the casting processes need to be monitored, evaluated and controlled to achieve expected quality levels.

The literature addresses quality improvement in foundry industry primarily focusing on surface roughness, mechanical properties, dimensional accuracy and defects in the cast parts and components which are often affected by numerous process variables. Primary data are collected from the experts working in sand and investment casting processes. The authors perform machine learning analysis of the data to model the quality parameters with appropriate process variables. Further, cluster analysis using k-means clustering method is performed to develop clusters of correlated process variables for sand and investment casting processes.

The authors identified primary process variables determining each quality parameter using machine learning approach. Quality parameters such as surface roughness, defects, mechanical properties and dimensional accuracy are represented by the identified sand-casting process variables accurately up to 83%, 83%, 100% and 83% and are represented by the identified investment-casting process variables accurately up to 100%, 67%, 67% and 100% respectively. Moreover, the prioritization of process variables in influencing the quality parameters is established which further helps the practitioners to monitor and control them within acceptable levels. Further the clusters of process variables help in analyzing their combined effect on quality parameters of casting products.

This study identified potential process variables and collected data from experts, researchers and practitioners on the effect of these on the quality aspects of cast products. While most of the previous studies focus on a very limited process variables for enhancing the quality characteristics of cast parts and components, this study represents each quality parameter as the function of influencing process variables which will enable the quality managers in Indian foundries to maintain capability and stability of casting processes. The models hence developed for both sand and investment casting for each quality parameter are validated with real life applications. Such studies are scarcely reported in the literature.

This study aims to develop a robust theoretical framework to explain the impact of hotels’ green brand authenticity on guests’ perceptions of well-being, customer engagement and approach behaviors toward green brands.

In this study, the authors examined the effect of green brand authenticity on perceptions of well-being, customer engagement and approach behaviors toward green brands. For the quantitative empirical analysis, 352 samples were used. Green brand authenticity integrates quality commitment, heritage, uniqueness and symbolism as high-dimensional factors.

The study conceptualizes green brand authenticity as a multi-dimensional phenomenon with four dimensions: quality commitment, heritage, uniqueness and symbolism. The results showed that green brand authenticity has a positive effect on hotel guests’ perceived well-being and behavioral intentions. Interestingly, environmental values did not have a statistically significant regulatory role, while green behavior in everyday life had a partial regulatory role.

This study aims to develop and empirically test a conceptual model that depicts the function of green authenticity in explaining customer responses to green brands. The results and the theoretical framework proposed in this study provide significant insights for researchers and practitioners in the hotel industry.

Further than evaluating brand authenticity generally, this study evaluates the authenticity of a brand's environmental protection efforts. As a result of the empirical analysis conducted in this study, the green brand authenticity of a hotel had a positive effect on customers’ emotional and behavioral aspects. This finding provided valuable and meaningful insights for green hotels and hotel brand-related research.

We aim to clarify the effect of economic uncertainty on Chinese stock market fluctuations. We extend the understanding of the asymmetric connectedness between economic uncertainty and stock market risk and provide different characteristics of spillovers from economic uncertainty to both upside and downside risk. Furthermore, we aim to provide the different impact patterns of stock market volatility following several exogenous shocks.

We construct a Chinese economic uncertainty index using a Factor-Augmented Variable Auto-Regressive Stochastic Volatility (FAVAR-SV) model for high-dimensional data. We then examine the asymmetric impact of realized volatility and economic uncertainty on the long-term volatility components of the stock market through the asymmetric Generalized Autoregressive Conditional Heteroskedasticity-Mixed Data Sampling (GARCH-MIDAS) model.

Negative news, including negative return-related volatility and higher economic uncertainty, has a greater impact on the long-term volatility components than positive news. During the financial crisis of 2008, economic uncertainty and realized volatility had a significant impact on long-term volatility components but did not constitute long-term volatility components during the 2015 A-share stock market crash and the 2020 COVID-19 pandemic. The two-factor asymmetric GARCH-MIDAS model outperformed the other two models in terms of explanatory power, fitting ability and out-of-sample forecasting ability for the long-term volatility component.

Many GARCH series models can also combine the GARCH series model with the MIDAS method, including but not limited to Exponential GARCH (EGARCH) and Threshold GARCH (TGARCH). These diverse models may exhibit distinct reactions to economic uncertainty. Consequently, further research should be undertaken to juxtapose alternative models for assessing the stock market response.

Our conclusions have important implications for stakeholders, including policymakers, market regulators and investors, to promote market stability. Understanding the asymmetric shock arising from economic uncertainty on volatility enables market participants to assess the potential repercussions of negative news, engage in timely and effective volatility prediction, implement risk management strategies and offer a reference for financial regulators to preemptively address and mitigate systemic financial risks.

First, in the face of domestic and international uncertainties and challenges, policymakers must increase communication with the market and improve policy transparency to effectively guide market expectations. Second, stock market authorities should improve the basic regulatory system of the capital market and optimize investor structure. Third, investors should gradually shift to long-term value investment concepts and jointly promote market stability.

This study offers a novel perspective on incorporating a Chinese economic uncertainty index constructed by a high-dimensional FAVAR-SV model into the asymmetric GARCH-MIDAS model.

For decades, financing constraints have been a major obstacle to corporate performance. Volumes have been written about the probable factors that can help firms alleviate such financial constraints. Nonetheless, empirical evidence concerning the various perspectives on how inventory control may influence financing constraints has been surprisingly scant. Using the resource- and region-based view as theoretical lenses, this study seeks to estimate the relationship between lean inventory, regional financial technology (fintech) and financing constraints.

Utilizing a large-scale sample of small- and medium-sized enterprises (SMEs) in China's manufacturing sector, the authors empirically test their hypotheses by using hierarchical linear regression models with multiple high-dimensional fixed effects.

Results indicate that firms with higher levels of inventory leanness and those located in more fintech-developed regions are less likely to encounter financing constraints. Furthermore, inventory leanness and regional fintech ecosystem development interact with each other to mitigate financing constraints. Moreover, inventory leanness significantly decreases firms' financing constraints when the regional fintech ecosystem is highly developed.

The present research contributes to the literature on the interface of supply chain management and financial management. It also provides managerial implications for policymakers and SME stakeholders.

The main goal of this research work is the optimization of the purchasing business process in the Moroccan public sector in terms of transparency and budgetary optimization. The authors have worked on the public university as an implementation field.

The design of the research work followed the design science research (DSR) methodology for information systems. DSR is a research paradigm wherein a designer answers questions relevant to human problems through the creation of innovative artifacts, thereby contributing new knowledge to the body of scientific evidence. The authors have adopted a techno-functional approach. The technical part consists of the development of an intelligent recommendation system that supports the choice of optimal information technology (IT) equipment for decision-makers. This intelligent recommendation system relies on a set of functional and business concepts, namely the Moroccan normative laws and Control Objectives for Information and Related Technology's (COBIT) guidelines in information system governance.

The modeling of business processes in public universities is established using business process model and notation (BPMN) in accordance with official regulations. The set of BPMN models constitute a powerful repository not only for business process execution but also for further optimization. Governance generally aims to reduce budgetary wastes, and the authors' recommendation system demonstrates a technical and methodological approach enabling this feature. Implementation of artificial intelligence techniques can bring great value in terms of transparency and fluidity in purchasing business process execution.

Business limitations: First, the proposed system was modeled to handle one type products, which are computer-related equipment. Hence, the authors intend to extend the model to other types of products in future works. Conversely, the system proposes optimal purchasing order and assumes that decision makers will rely on this optimal purchasing order to choose between offers. In fact, as a perspective, the authors plan to work on a complete automation of the workflow to also include vendor selection and offer validation. Technical limitations: Natural language processing (NLP) is a widely used sentiment analysis (SA) technique that enabled the authors to validate the proposed system. Even working on samples of datasets, the authors noticed NLP dependency on huge computing power. The authors intend to experiment with learning and knowledge-based SA and assess the' computing power consumption and accuracy of the analysis compared to NLP. Another technical limitation is related to the web scraping technique; in fact, the users' reviews are crucial for the authors' system. To guarantee timeliness and reliable reviews, the system has to look automatically in websites, which confront the authors with the limitations of the web scraping like the permanent changing of website structure and scraping restrictions.

The modeling of business processes in public universities is established using BPMN in accordance with official regulations. The set of BPMN models constitute a powerful repository not only for business process execution but also for further optimization. Governance generally aims to reduce budgetary wastes, and the authors' recommendation system demonstrates a technical and methodological approach enabling this feature.

The adopted techno-functional approach enabled the authors to bring information system governance from a highly abstract level to a practical implementation where the theoretical best practices and guidelines are transformed to a tangible application.

This study aims to discuss the state-of-the-art digital factory (DF) development combining digital twins (DTs), sensing devices, laser additive manufacturing (LAM) and subtractive manufacturing (SM) processes. The current shortcomings and outlook of the DF also have been highlighted. A DF is a state-of-the-art manufacturing facility that uses innovative technologies, including automation, artificial intelligence (AI), the Internet of Things, additive manufacturing (AM), SM, hybrid manufacturing (HM), sensors for real-time feedback and control, and a DT, to streamline and improve manufacturing operations.

This study presents a novel perspective on DF development using laser-based AM, SM, sensors and DTs. Recent developments in laser-based AM, SM, sensors and DTs have been compiled. This study has been developed using systematic reviews and meta-analyses (PRISMA) guidelines, discussing literature on the DTs for laser-based AM, particularly laser powder bed fusion and direct energy deposition, in-situ monitoring and control equipment, SM and HM. The principal goal of this study is to highlight the aspects of DF and its development using existing techniques.

A comprehensive literature review finds a substantial lack of complete techniques that incorporate cyber-physical systems, advanced data analytics, AI, standardized interoperability, human–machine cooperation and scalable adaptability. The suggested DF effectively fills this void by integrating cyber-physical system components, including DT, AM, SM and sensors into the manufacturing process. Using sophisticated data analytics and AI algorithms, the DF facilitates real-time data analysis, predictive maintenance, quality control and optimal resource allocation. In addition, the suggested DF ensures interoperability between diverse devices and systems by emphasizing standardized communication protocols and interfaces. The modular and adaptable architecture of the DF enables scalability and adaptation, allowing for rapid reaction to market conditions.

Based on the need of DF, this review presents a comprehensive approach to DF development using DTs, sensing devices, LAM and SM processes and provides current progress in this domain.