Search results

Intent detection (ID) and slot filling (SF) are two important tasks in natural language understanding. ID is to identify the main intent of a paragraph of text. The goal of SF is to extract the information that is important to the intent from the input sentence. However, most of the existing methods use sentence-level intention recognition, which has the risk of error propagation, and the relationship between intention recognition and SF is not explicitly modeled. Aiming at this problem, this paper proposes a collaborative model of ID and SF for intelligent spoken language understanding called ID-SF-Fusion.

ID-SF-Fusion uses Bidirectional Encoder Representation from Transformers (BERT) and Bidirectional Long Short-Term Memory (BiLSTM) to extract effective word embedding and context vectors containing the whole sentence information respectively. Fusion layer is used to provide intent–slot fusion information for SF task. In this way, the relationship between ID and SF task is fully explicitly modeled. This layer takes the result of ID and slot context vectors as input to obtain the fusion information which contains both ID result and slot information. Meanwhile, to further reduce error propagation, we use word-level ID for the ID-SF-Fusion model. Finally, two tasks of ID and SF are realized by joint optimization training.

We conducted experiments on two public datasets, Airline Travel Information Systems (ATIS) and Snips. The results show that the Intent ACC score and Slot F1 score of ID-SF-Fusion on ATIS and Snips are 98.0 per cent and 95.8 per cent, respectively, and the two indicators on Snips dataset are 98.6 per cent and 96.7 per cent, respectively. These models are superior to slot-gated, SF-ID NetWork, stack-Prop and other models. In addition, ablation experiments were performed to further analyze and discuss the proposed model.

This paper uses word-level intent recognition and introduces intent information into the SF process, which is a significant improvement on both data sets.

With the rapid development of short videos in China, the public has become accustomed to using short videos to express their opinions. This paper aims to solve problems such as how to represent the features of different modalities and achieve effective cross-modal feature fusion when analyzing the multi-modal sentiment of Chinese short videos (CSVs).

This paper aims to propose a sentiment analysis model MSCNN-CPL-CAFF using multi-scale convolutional neural network and cross attention fusion mechanism to analyze the CSVs. The audio-visual and textual data of CSVs themed on “COVID-19, catering industry” are collected from CSV platform Douyin first, and then a comparative analysis is conducted with advanced baseline models.

The sample number of the weak negative and neutral sentiment is the largest, and the sample number of the positive and weak positive sentiment is relatively small, accounting for only about 11% of the total samples. The MSCNN-CPL-CAFF model has achieved the Acc-2, Acc-3 and F1 score of 85.01%, 74.16 and 84.84%, respectively, which outperforms the highest value of baseline methods in accuracy and achieves competitive computation speed.

This research offers some implications regarding the impact of COVID-19 on catering industry in China by focusing on multi-modal sentiment of CSVs. The methodology can be utilized to analyze the opinions of the general public on social media platform and to categorize them accordingly.

This paper presents a novel deep-learning multimodal sentiment analysis model, which provides a new perspective for public opinion research on the short video platform.

In this study, the authors examine artificial knowledge as a fundamental stream of knowledge management for sustainable and resilient business models in supply chain management (SCM). The study aims to provide a comprehensive overview of artificial knowledge and digitalization as key enablers of the improvement of SCM accountability and sustainable performance towards the UN 2030 Agenda.

Using the SCOPUS database and Google Scholar, the authors analyzed 135 English-language publications from 1990 to 2022 to chart the pattern of knowledge production and dissemination in the literature. The data were collected, reviewed and peer-reviewed before conducting bibliometric analysis and a systematic literature review to support future research agenda.

The results highlight that artificial knowledge and digitalization are linked to the UN 2030 Agenda. The analysis further identifies the main issues in achieving sustainable and resilient SCM business models. Based on the results, the authors develop a conceptual framework for artificial knowledge and digitalization in SCM to increase accountability and sustainable performance, especially in times of sudden crises when business resilience is imperative.

The study results add to the extant literature by examining artificial knowledge and digitalization from the resilience theory perspective. The authors suggest that different strategic perspectives significantly promote resilience for SCM digitization and sustainable development. Notably, fostering diverse peer exchange relationships can help stimulate peer knowledge and act as a palliative mechanism that builds digital knowledge to strengthen and drive future possibilities.

This research offers valuable guidance to supply chain practitioners, managers and policymakers in re-thinking, re-formulating and re-shaping organizational processes to meet the UN 2030 Agenda, mainly by introducing artificial knowledge in digital transformation training and education programs. In doing so, firms should focus not simply on digital transformation but also on cultural transformation to enhance SCM accountability and sustainable performance in resilient business models.

This study is, to the authors' best knowledge, among the first to conceptualize artificial knowledge and digitalization issues in SCM. It further integrates resilience theory with institutional theory, legitimacy theory and stakeholder theory as the theoretical foundations of artificial knowledge in SCM, based on firms' responsibility to fulfill the sustainable development goals under the UN's 2030 Agenda.

The relevance of analytics to the healthcare supply chain is increasing with emerging trends and technologies. This study examines how analytics are used in the healthcare supply chain in the “new normal” environment.

A systematic literature review was conducted by extracting research articles related to analytics in the healthcare supply chain from Scopus. The author used a hybrid review approach that combines bibliometric analysis with a theories, contexts, characteristics, and methodology (TCCM) framework-based review to identify various themes of analytics in the healthcare supply chain.

The hybrid review strategy yielded results that focus on prevalent theories, contexts, characteristics, and methodologies in the field of healthcare supply chain analytics. Future research should explore the resulting antecedents, decision-making processes and outcomes (ADO) framework, which integrates technological, economic, and societal concerns and outcomes. Future research agendas could also seek to apply theoretical perspectives in the field of analytics in the healthcare supply chain.

The result of a review of selected studies adds to the current body of work and contributes to the growth of research in the field of analytics in the healthcare supply chain. It also provides new directions to healthcare supply chain managers and academic scholars.

The purpose of this paper is to improve the existing paradigm of edge computing to maintain a balanced energy usage.

The new greedy algorithm is proposed to balance the energy consumption in edge computing.

The new greedy algorithm can balance energy more efficiently than the random approach by an average of 66.59 percent.

The results are shown in this paper which are better as compared to existing algorithms.

In recent years, there has been a growing interest in the potential of data analytics to enhance project delivery. Yet many argue that its application in projects is still lagging behind other disciplines. This paper aims to provide a review of the current use of data analytics in project delivery encompassing both academic research and practice to accelerate current understanding and use this to formulate questions and goals for future research.

We propose to achieve the research aim through the creation of a systematic review of the status of data analytics in project delivery. Fusing the methodology of integrative literature review with a recently established practice to include both white and grey literature amounts to an approach tailored to the state of the domain. It serves to delineate a research agenda informed by current developments in both academic research and industrial practice.

The literature review reveals a dearth of work in both academic research and practice relating to data analytics in project delivery and characterises this situation as having “more gap than knowledge.” Some work does exist in the application of machine learning to predicting project delivery though this is restricted to disparate, single context studies that do not reach extendible findings on algorithm selection or key predictive characteristics. Grey literature addresses the potential benefits of data analytics in project delivery but in a manner reliant on “thought-experiments” and devoid of empirical examples.

Based on the review we articulate a research agenda to create knowledge fundamental to the effective use of data analytics in project delivery. This is structured around the functional framework devised by this investigation and highlights both organisational and data analytic challenges. Specifically, we express this structure in the form of an “onion-skin” model for conceptual structuring of data analytics in projects. We conclude with a discussion about if and how today’s project studies research community can respond to the totality of these challenges. This paper provides a blueprint for a bridge connecting data analytics and project management.

Despite the growing attention on the relevance of improved building management systems with cognition in recent years in the architecture, engineering, construction and operation (AECO) community, no review has been conducted to understand the human-environment interaction features of cyber-physical systems (CPS) and digital twins (DTs) in developing the concept of a cognitive building (CB). Thus, this paper aims to review existing studies on CPS and DTs for CB to propose a comprehensive system architecture that considers human-environment interactions.

Scientometric analysis and content analysis were adopted for this study.

The scientometric analysis of 1,042 journal papers showed the major themes of CPS/DTs for CB, and these can be categorized into three key technologies to realize CB in the AECO community: CPS, DTs and cognitive computing (CC). Content analysis of 44 relevant publications in the built environment assisted in understanding and evidently confirming the claim of this study on the integration of CPS and DTs for CB in construction by also involving the CC. It is found and confirmed that CB can be realized with CPS and DTs along with the CC. A CB system architecture (CBSA) is proposed from the three key technologies considering the human-environment interactions in the loop. The study discovered the potential applications of the CBSA across the building lifecycle phases, including the design, construction and operations and maintenance, with the potential promise of endowing resilience, intelligence, greater efficiency and self-adaptiveness. Based on the findings of the review, four research directions are proposed: human-environment interactions, CB for sustainable building performance, CB concept for modular buildings and moving beyond CB.

This study stands out for comprehensively surveying the intellectual core and the landscape of the general body of knowledge on CPS/DTs for CB in the built environment. It makes a distinctive contribution to knowledge as it does not only propose CBSA by integrating CPS and DTs along with CC but also suggests some potential practical applications. These may require expert judgments and real case examples to enhance reproducibility and validation.

Fall accidents can cause casualties and economic losses in the construction industry. Fall portents, such as loss of balance (LOB) and sudden sways, can result in fatal, nonfatal or attempted fall accidents. All of them are worthy of studying to take measures to prevent future accidents. Detecting fall portents can proactively and comprehensively help managers assess the risk to workers as well as in the construction environment and further prevent fall accidents.

This study focused on the postures of workers and aimed to directly detect fall portents using a computer vision (CV)-based noncontact approach. Firstly, a joint coordinate matrix generated from a three-dimensional pose estimation model is employed, and then the matrix is preprocessed by principal component analysis, K-means and pre-experiments. Finally, a modified fusion K-nearest neighbor-based machine learning model is built to fuse information from the x, y and z axes and output the worker's pose status into three stages.

The proposed model can output the worker's pose status into three stages (steady–unsteady–fallen) and provide corresponding confidence probabilities for each category. Experiments conducted to evaluate the approach show that the model accuracy reaches 85.02% with threshold-based postprocessing. The proposed fall-portent detection approach can extract the fall risk of workers in the both pre- and post-event phases based on noncontact approach.

First, three-dimensional (3D) pose estimation needs sufficient information, which means it may not perform well when applied in complicated environments or when the shooting distance is extremely large. Second, solely focusing on fall-related factors may not be comprehensive enough. Future studies can incorporate the results of this research as an indicator into the risk assessment system to achieve a more comprehensive and accurate evaluation of worker and site risk.

The proposed machine learning model determines whether the worker is in a status of steady, unsteady or fallen using a CV-based approach. From the perspective of construction management, when detecting fall-related actions on construction sites, the noncontact approach based on CV has irreplaceable advantages of no interruption to workers and low cost. It can make use of the surveillance cameras on construction sites to recognize both preceding events and happened accidents. The detection of fall portents can help worker risk assessment and safety management.

Existing studies using sensor-based approaches are high-cost and invasive for construction workers, and others using CV-based approaches either oversimplify by binary classification of the non-entire fall process or indirectly achieve fall-portent detection. Instead, this study aims to detect fall portents directly by worker's posture and divide the entire fall process into three stages using a CV-based noncontact approach. It can help managers carry out more comprehensive risk assessment and develop preventive measures.

Monitoring of the quality of precast concrete (PC) components is crucial for the success of prefabricated construction projects. Currently, quality monitoring of PC components during the construction phase is predominantly done manually, resulting in low efficiency and hindering the progress of intelligent construction. This paper presents an intelligent inspection method for assessing the appearance quality of PC components, utilizing an enhanced you look only once (YOLO) model and multi-source data. The aim of this research is to achieve automated management of the appearance quality of precast components in the prefabricated construction process through digital means.

The paper begins by establishing an improved YOLO model and an image dataset for evaluating appearance quality. Through object detection in the images, a preliminary and efficient assessment of the precast components' appearance quality is achieved. Moreover, the detection results are mapped onto the point cloud for high-precision quality inspection. In the case of precast components with quality defects, precise quality inspection is conducted by combining the three-dimensional model data obtained from forward design conversion with the captured point cloud data through registration. Additionally, the paper proposes a framework for an automated inspection platform dedicated to assessing appearance quality in prefabricated buildings, encompassing the platform's hardware network.

The improved YOLO model achieved a best mean average precision of 85.02% on the VOC2007 dataset, surpassing the performance of most similar models. After targeted training, the model exhibits excellent recognition capabilities for the four common appearance quality defects. When mapped onto the point cloud, the accuracy of quality inspection based on point cloud data and forward design is within 0.1 mm. The appearance quality inspection platform enables feedback and optimization of quality issues.

The proposed method in this study enables high-precision, visualized and automated detection of the appearance quality of PC components. It effectively meets the demand for quality inspection of precast components on construction sites of prefabricated buildings, providing technological support for the development of intelligent construction. The design of the appearance quality inspection platform's logic and framework facilitates the integration of the method, laying the foundation for efficient quality management in the future.

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.