Search

The paper reports findings from a study that examining how cross-platform social media analysis can help to map the digital discourse on sustainable mobility and sustainable transport, and enhance the understanding of sociotechnical low-carbon transport transitions. Using the hashtag search queries #sustainabletransport and #sustainablemobility, 33,121 Tweets (2013–2021) and 8,089 Instagram images including captions (2017/2018–2021) were scraped using the Python modules Twint and Instaloader. Quantitative text and sentiment analyses were applied to the Tweets and image captions. Additionally, an automated machine learning-based image analysis of the Instagram images was conducted using object detection via OpenCV. Synthesized results formed the base for a cross-platform analysis inspired by Rogers' method comprising hot topics/key themes, user mentions, sentiment polarity, and co-hashtags. Notably, electromobility emerged as a prominent theme, particularly on Instagram, while #sustainabletransport was closely associated with active travel, notably bicycling, and #sustainablemobility showcased a dominance of electromobility discourse. The study demonstrates the investigative potentials of cross-platform social media analysis studies to enhance the understanding of sociotechnical low-carbon transport transitions. Drawing on key results, the paper suggests an adapted version of the Geelsean Multi-Level Perspective on Sociotechnical Transitions. [ABSTRACT FROM AUTHOR]

Developing and executing large-scale data analysis applications in parallel and distributed environments can be a complex and time-consuming task. Developers often find themselves diverted from their application logic to handle technical details about the underlying runtime and related issues. To simplify this process, ParSoDA, a Java library, has been proposed to facilitate the development of parallel data mining applications executed on HPC systems. It simplifies the process by providing built-in scalability mechanisms relying on the Hadoop and Spark frameworks. This paper presents ParSoDA-Py, the Python version of the ParSoDA library, which allows for further support of commonly used runtimes and libraries for big data analysis. After a complete library redesign, ParSoDA can be now easily integrated with other Python-based distributed runtimes for HPC systems, such as COMPSs and Apache Spark, and with the large ecosystem of Python-based data processing libraries. The paper discusses the adaptation process, which takes into consideration the new technical requirements, and evaluates both usability and scalability through some case study applications. [ABSTRACT FROM AUTHOR]

This paper aims to understand the impact of big data analytics on the retail supply chain. For doing so, we set our context to select the best big data practices amongst the available alternatives based on retail supply chain performance. We have applied TODIM (an acronym in Portuguese for Interactive Multi-criteria Decision Making) for the selection of the best big data analytics tools among the identified nine practices (data science, neural networks, enterprise resource planning, cloud computing, machine learning, data mining, RFID, Blockchain and IoT and Business intelligence) based on seven supply chain performance criteria (supplier integration, customer integration, cost, capacity utilization, flexibility, demand management, and time and value). One of the intriguing understandings from this paper is that most of the retail firms are in a dilemma between customer loyalty and cost while implementing the big data practices in their organization. This study analyses the dominance of the big data practices at the retail supply chain level. This helps the newly emerging retail firms in evaluating the best big data practice based on the importance and dominance of supply chain performance measures. [ABSTRACT FROM AUTHOR]

This research paper focuses on the challenges associated with learning classifiers from large-scale, highly imbalanced datasets prevalent in many real-world applications. Traditional algorithms learning often need better performance and high computational efficiency when dealing with imbalanced data. Factors such as class imbalance, noise, and class overlap make it demanding to learn effective classifiers. In this study, we propose a novel self-paced ensemble framework for classifying imbalanced data. The framework employs under-sampling to self-harmonize data hardness and build a robust ensemble. Extensive experimental testing demonstrates promising results in handling overlapping classes and skewed distributions while maintaining computational efficiency. The self-paced ensemble method addresses the challenges of high imbalance ratios, class overlap, and noise presence in large-scale imbalanced classification problems. By incorporating the knowledge of these challenges into our learning framework, we establish the concept of classification hardness distribution, and the self-paced ensemble is a revolutionary learning paradigm for massive imbalance categorization, capable of improving the performance of existing learning algorithms on imbalanced data and providing better results for future applications. [ABSTRACT FROM AUTHOR]

Classifying imbalanced data presents a significant challenge, and many studies have proposed methodologies to address this issue. Among them, Convolutional Neural Networks have demonstrated superior performance for imbalanced image classification. This paper initially employs various data pre-processing methods such as over-sampling, under-sampling, and SMOTE to enhance the original dataset. Subsequently, an Ensemble CNN learning model is used to train and predict the data. In order to comprehensively evaluate models trained on imbalanced data, we used metrics such as Accuracy, Recall, Precision, F1-score, and G-mean. On the CIFAR-10 and Fashion-MNIST datasets, different samples from each category were extracted as imbalanced data for experimental research. Compared to the AdaBoost-DenseNet model, our proposed methodology increases the test accuracy on the CIFAR-10 dataset by 9%. Similarly, the F1-score and G-mean improved by 0.096 and 0.069, respectively. Compared to traditional methodologies, our proposed method significantly improves accuracy, recall, precision, and other performance indicators. [ABSTRACT FROM AUTHOR]

With the rapid development of AI and big data mining technologies, computerized medical decision-making has become increasingly prominent. The aim of high-utility pattern mining (HUPM) is to discover meaningful patterns in medical databases that contribute to maximizing the utility from the perspective of diagnosis. However, HUPM pays less attention to the interpretability and explainability of these patterns in medical decision-making scenarios. This paper proposes a novel algorithm called the Improved fuzzy high-utility pattern mining (IF-HUPM) to address this problem. First, the paper applies a fuzzy preprocessing method to divide the fuzzy intervals of a medical quantitative data set, which enhances the fuzziness and interpretability of the data. Next, in the process of IF-HUPM, both fuzzy tree and list structures are employed to calculate fuzzy high-utility values. By combining the characteristics of the one-stage and two-stage algorithms of HUPM, an adaptive-phase Fuzzy HUPM hybrid frame is proposed. The experimental results demonstrate that the proposed IF-HUPM algorithm enhances both accuracy and efficiency and the mining process requires less time and space on average. [ABSTRACT FROM AUTHOR]

The location planning of relief distribution centres (DCs) is crucial in humanitarian logistics as it directly influences the disaster response and service to the affected victims. In light of the critical role of facility location in humanitarian logistics planning, the study proposes a two-stage relief distribution location problem. The first stage of the model determines the minimum number of relief DCs, and the second stage find the optimal location of these DCs to minimize the total cost. To address a more realistic situation, restrictions are imposed on the coverage area and capacity of each DCs. In addition, for optimally solving this complex NP-hard problem, a novel two-phase algorithm with exploration and exploitation phase is developed in the paper. The first phase of the algorithm i.e., exploration phase identifies a near-optimal solution while the second phase i.e. exploitation phase enhances the solution quality through a close circular proximity investigation. Furthermore, the comparative analysis of the proposed algorithm with other well-known algorithms such as genetic algorithm, pattern search, fmincon, multistart and hybrid heuristics is also reported and computationally tested from small to large data sets. The results reveal that the proposed two-phase algorithm is more efficient and effective when compared to the conventional metaheuristic methods. [ABSTRACT FROM AUTHOR]

Longitudinal data, measurements taken from the same subjects over time, appear routinely in many scientific fields, such as biomedical science, public health, ecology and environmental sciences. With the rapid development of information technology, modern longitudinal data are becoming massive in volume and high dimensional, hence often require distributed analysis in real-world applications. Standard divide-and-conquer techniques do not apply directly to longitudinal big data due to within-subject dependence. In this paper, we focus on developing a distributed algorithm to support quantile regression (QR) analysis of longitudinal big data, which currently remains an open and challenging issue. We employ weighted quantile regression (WQR) to accommodate the correlation in longitudinal big data, and parallelize the WQR estimation process with a two-stage algorithm to support distributed computing. Based on weights estimated in the first stage by the Newton–Raphson algorithm, the second stage solves the WQR problem using the multi-block alternating direction method of multipliers (ADMM). Simulation studies show that, compared to traditional non-distributed algorithms, our proposed method has favorable estimation accuracy and is computationally more efficient in both non-distributed and distributed environments. Further, we also analyze an air quality data set to illustrate the practical performance of this method. [ABSTRACT FROM AUTHOR]

Choice architecture describes the environment in which choices are presented to decision-makers. In recent years, public and private actors have looked at choice architecture with great interest as they seek to influence human behaviour. These actors are typically called choice architects. Increasingly, however, this role of architecting choice is not performed by a human choice architect, but an algorithm or artificial intelligence, powered by a stream of Big Data and infused with an objective it has been programmed to maximise. We call this entity the autonomous choice architect. In this paper, we present an account of why artificial intelligence can fulfil the role of a choice architect and why this creates problems of transparency, responsibility and accountability for nudges. We argue that choice architects, be them autonomous computational systems or human-beings, at a most basic level select, from a range of designs, the design which is most likely to maximise a pre-determined objective. We then proceed to argue that, given the growing demand for targeted, personalised choice architecture and for faster, dynamic reconfigurations of choice architecture, as well as the ever-expanding pool of data from which feedback can be drawn, the role of the human choice architect is increasingly obscured behind algorithmic, artificially intelligent systems. We provide a discussion of the implications of autonomous choice architects, focusing on the importance of the humans who programme these systems, ultimately arguing that despite technological advances, the responsibility of choice architecture and influence remains firmly one human beings must bear. [ABSTRACT FROM AUTHOR]

In recent years, various studies have been conducted on SVMs and their applications in different area. They have been developed significantly in many areas. SVM is one of the most robust classification and regression algorithms that plays a significant role in pattern recognition. However, SVM has not been developed significantly in some areas like large-scale datasets, unbalanced datasets, and multiclass classification. Efficient SVM training in large-scale datasets is of great importance in the big data era. However, as the number of samples increases, the time and memory required to train SVM increase, making SVM impractical even for a medium-sized problem. With the emergence of big data, this problem becomes more significant. This paper presents a novel distributed method for SVM training in which a very small subset of training samples is used for classification, which reduces the problem size and thus the required memory and computational resources. The solution of this problem almost converges to standard SVM. This method includes three steps: first, detecting a subset of distributed training samples, second, creating local models of SVM and obtaining partial vectors, and finally combining the partial vectors and obtaining the global vector and the final model. In addition, the datasets which suffer from unbalanced number of samples and tend to the majority class, the proposed method balances the samples of the two classes and it can be used in unbalanced datasets. The empirical results show that using this method is efficient for large-scale problems. [ABSTRACT FROM AUTHOR]

The evolution of Internet technology has led to an increase in online users. This study focuses on the pivotal role of visual elements in web content conveyance and their impact on user browsing behavior. Therefore, the use of visual elements in web design based on big data has aroused widespread concern among web designers, they apply visual elements to their web design works to make the web more attractive. This study examines the composition and distribution characteristics of key visual elements identified through user behavior data in a big data environment and discusses the use of visual elements in web design in the era of network economy. In addition, this paper issued 200 questionnaires to investigate the degree of attention to visual elements in web pages for users of different occupations and different educational backgrounds. Our survey indicated that visual elements captured the attention of 41% of corporate employees, whereas a mere 1% of social welfare workers focused on web content; 36% of undergraduates pay attention to visual elements of web pages, but only 5% and 4% of postgraduates and doctoral degrees and above. Therefore, the visual elements of the designed web page need to conform to the user's cultural background and professional background. [ABSTRACT FROM AUTHOR]

Master–slave blockchain is a novel information processing technology that is domain-oriented and uses efficient cryptography principles for trustworthy communication and storage of big data. Existing indexing methods primarily target the creation of a single-structured blockchain, resulting in extensive time and memory requirements. As the scale of domain data continues to grow exponentially, master–slave blockchain systems face increasingly severe challenges with regards to low query efficiency and extended traceback times. To address these issues, this paper propose a multi-level index construction method for the master–slave blockchain (MLI). Firstly, MLI introduces a weight matrix and partitions the entire master–slave blockchain based on the master chain structure, the weight of each partition is assigned. Secondly, for the master blockchain in each partition, a master chain index construction method based on jump consistent hash (JHMI) is proposed, which takes the key value of the nodes and the number of index slots as input and outputs the master chain index. Finally, a bloom filter is introduced to improve the column-based selection function and build a secondary composite index on the subordinate blockchain corresponding to each master block. Experimental results on three constraint conditions and two types of datasets demonstrate that the proposed method reduce the index construction time by an average of 9.28%, improve the query efficiency by 12.07%, and reduce the memory overhead by 24.4%. [ABSTRACT FROM AUTHOR]

Green approaches remain little disseminated in the healthcare sector despite growing interest in recent years from practitioners and researchers. Big Data Analytics Capability (BDAC) can play a critical role in the integration of environmental concerns into operations and supply chain management (OSCM) and further strengthen the environmental performance of healthcare facilities. According to the literature, the integration of the environment into operations process remains insufficient to achieve high levels of performance and requires efforts in green process innovation. However, this relationship between BDAC and green process innovation remains poorly justified empirically. To address this theoretical gap, we investigated the relationship between BDAC, environmental process integration, green process innovation in OSCM and environmental performance. The main contribution of this study is the valuable knowledge on how BDAC influences environmental process integration and green process innovation to enhance environmental performance. Moreover, the study highlights the mediating role of green process innovation on environmental performance, a finding that has not been mentioned in the extant literature. The paper provides valuable insight for managers and stakeholders that can assist them in supporting the application of BDAC in healthcare OSCM to create sustainable value. [ABSTRACT FROM AUTHOR]

Data-driven innovation enables firms to design products that are more responsive to market needs, which greatly reduces the risk of innovation. Customer data in the same supply chain has certain commonality, but data separation makes it difficult to maximize data value. The selection of an appropriate mode for cooperation innovation should be based on the particular big data analytics capability of the firms. This paper focuses on the influence of big data analytics capability on the choice of cooperation mode, and the influence of their matching relationship on cooperation performance. Specifically, using game-theoretic models, we discuss two cooperation modes, data analytics is implemented individually (i.e., loose cooperation) by either firm, or jointly (tight cooperation) by both firms, and further discuss the addition of coordination contracts under the loose mode. Several important conclusions are obtained. Firstly, both firms' big data capability have positive effects on the selection of tight cooperation mode. Secondly, with the improvement of big data capability, the firms' innovative performance gaps between loose and tight mode will increase significantly. Finally, when the capability meet certain condition, the cost subsidy contract can alleviate the gap between the two cooperative models. [ABSTRACT FROM AUTHOR]

Many organizations are increasingly investing in building dynamic capabilities to gain competitive advantage. New products play an important role in gaining competitive advantage and can significantly boost organizational performance. Although new product development (NPD) is widely recognized as a potentially vital source of competitive advantage, organizations face challenges in terms of developing the right antecedents or capabilities to influence NPD performance. Our research suggests that organizations should invest in building alliance management capability (AMC), big data analytics capability (BDAC) and information visibility (IV) to achieve their desired NPD success. Informed by the dynamic capabilities view of the firm (DCV) we have stated seven research hypotheses. We further tested our hypotheses using 219 usable respondents gathered using a pre-tested instrument. The hypotheses were tested using variance based structural equation modelling (PLS-SEM). The results of our study paint an interesting picture. Our study makes some significant contribution to the DCV and offers some useful directions to practitioners engaged in NPD in the big data analytics era. We demonstrate that AMC and BDAC are lower-order dynamic capabilities and that AMC has a positive and significant influence on BDAC. In turn, AMC and BDAC influence NPD under the moderating influence of IV. Ours is one of the first studies to empirically establish an association among three distinct dynamic capabilities which are often considered in isolation: AMC, BDAC and NPD. Our findings support emergent views on dynamic capabilities and their classification into various orders. Lastly, we provide empirical evidence that information visibility acts as a contingent variable to both AMC and BDAC effects on NPD. We end our paper by outlining some limitations of our study and by offering useful future research directions. [ABSTRACT FROM AUTHOR]

The Cross-Taiwan Strait Railway (CTSR) is a significant construction, and the automatic driving of high-speed trains (HSTs) on the CTSR is a crucial technology for its operation. Before opening the CTSR, it is forward-looking to conduct theoretical research. Inspired by AlphaZero, this paper combines expert experience and Newtonian mechanics to propose a method on automatically generating massive virtual driving curves for HSTs of the CTSR. To compress the solution space from infinite to finite, we set the driving acceleration and speed limit boundary of HSTs from the perspective of expert experience. The model then combines big data technology to generate massive virtual driving curves automatically and uses statistical analysis to select high-performance ones. We found that 1) numerous virtual driving curves can be automatically generated according to different restricted speeds and line conditions, with solid adaptability; 2) the driving curves cover a specific running time and reach all running times in the range, with good ergodicity; 3) real-time tracking of the running time of each driving curve by 0.1 s, with high accuracy; 4) lots of driving curves are available on each run time for selecting curves with best performance, with good selectivity. In conclusion, this method can automatically generate virtual driving curves for HSTs on the CTSR, providing high-quality virtual data for future research on the automatic driving of HSTs on the CTSR. [ABSTRACT FROM AUTHOR]

Although there are numerous data modeling tools for relational databases, data modeling for NoSQL databases has seen another perspective. These databases (a) do not define any explicit schema, (b) store data in a denormalized manner, and (c) give many structure alternatives. The decision to structure the data always relies on rules of thumb, which do not guarantee an optimal structural solution. Based on this motivation, this paper offers a workload-driven model for the logical schema design of a NoSQL document database. It consists of Model input, Intermediate transformation, and Final schema generation. The proposed model takes the conceptual schema (EER model) and application workload (queries and anticipated data volume) as input and describes a procedure to convert it into a logical model for NoSQL document stores. The conversion process initially converts the application queries into query graphs. The query graphs, along with the anticipated data volume, are used to generate the query labels. The resulting query labels are assigned on the schema graph designed from the EER model. The schema graph and labels are used to transform the EER model into the appropriate logical schema model based on the actions defined for each label. We evaluate the model using a case study in the eCommerce application domain. The experimental evaluation shows the proposed model outperforms the existing conventional, optimized, and query path graphs models in multiple aspects, including query performance, storage space efficiency, aggregate pipeline efficiency, read–write latency, collection-wise performance, scalability, throughput and latency. By effectively addressing the challenges associated with managing the variety and volume of big data through a well-designed schema, our proposed model significantly reduces the time, cost, and effort required for schema development and repair. [ABSTRACT FROM AUTHOR]

Due to the intricate mechanism of Abbe error's impact on spatial accuracy, the accumulated Abbe error in the traditional spatial accuracy model is difficult to identify and eliminate, resulting in reduced modeling accuracy and limiting accuracy improvement. This paper proposes a data-driven spatial accuracy modeling approach for machine tools under the influence of Abbe error, using a three-axis coupling measurement optical path to directly measure the comprehensive spatial accuracy data of the machine tool containing Abbe error. Firstly, to effectively identify Abbe error in the comprehensive spatial accuracy, an Abbe error quantization function is established, which analyzes its formation mechanism in the measurement process to eliminate Abbe error in the machine tool's spatial accuracy data. Furthermore, to address the issue of limited data samples after eliminating Abbe error, data samples are extended based on the degradation mechanism of machine tool spatial accuracy at different coordinate positions, resulting in a high-precision spatial error model for machine tools. Finally, the experiment is conducted on a three-axis CNC machine tool, with a model accuracy of over 95%. The example application verification demonstrates that the developed model scheme is feasible and effective. [ABSTRACT FROM AUTHOR]

Businesses nowadays often feel impelled to rush and implement Big Data and Artificial Intelligence initiatives in their organizations without clarity on their business problems, nor the appropriate methodologies for extracting actionable insights from the data. In contrast, this paper presents a process that starts with an articulated business problem instead of a "data rush". The presented process, dubbed the Business-Driven Data-Supported (BDDS) process, is rigorously anchored in concepts from Theory of Constraints and Information Theory. BDDS guides businesses in solving their problems by stating observed performance gaps, uncovering their underlying root cause, formulating the "right question", utilizing only the "right data", and choosing the "right methodology" to extract the "right information" from the data, leading to the "right actionable insights." Also provided is a prescriptive framework, dubbed the Data-to-Information-Extraction-Methodology (DIEM), for selecting the "right methodology", based on the understanding level of relevant system dependencies and the availability of relevant data. The BDDS process is illustrated by an example from the healthcare industry, and the efficacy and applicability of the DIEM framework are supported by eleven case studies from a broad range of industries. [ABSTRACT FROM AUTHOR]

This paper presents a multi-modal recognition system (MMR) which can eliminates the need for using Barcodes and Radio Frequency Identification Systems (RFIDs). Barcodes and RFID have limitations; for example, Barcodes require the scanner to have a direct line of sight with the code, and they are more susceptible to errors. Barcodes can also be hard to locate and can be affixed to oddly shaped products. While, RFID may overcome such problems, but it could be disturbed if the RFID is attached to a metallic background. The proposed MMR system can monitor items flowing one by one down on a conveyor belt to make sure they match their images, thus enabling robotics identification for items while picking them up, sorting them or turning them into a desirable orientation. Recognizing the business landscape evolves and competition from low-cost nations grows, new models must be created that provide a competitive edge by combining the Lean paradigm with Industry 4.0 technical advancements. This paper reports a contribution to this field by assessing the supporting function of MMR state-of-the-art algorithms in Lean manufacturing. In addition, this paper also aims to explore how MMR could be integrated into the Lean manufacturing settings to enable a competitive manufacturing process in a Lean 4.0 environment. A dataset with 21,000 vegetable images of 15 classes was used to present the recent development and application of image analysis and computer vision systems in object recognition, showing an overall detection F1-score of 85.08%. [ABSTRACT FROM AUTHOR]

Gene co-expression networks are valuable tools for discovering biologically relevant information within gene expression data. However, analysing large datasets presents challenges due to the identification of nonlinear gene–gene associations and the need to process an ever-growing number of gene pairs and their potential network connections. These challenges mean that some experiments are discarded because the techniques do not support these intense workloads. This paper presents pyEnGNet, a Python library that can generate gene co-expression networks in High-performance computing environments. To do this, pyEnGNet harnesses CPU and multi-GPU parallel computing resources, efficiently handling large datasets. These implementations have optimised memory management and processing, delivering timely results. We have used synthetic datasets to prove the runtime and intensive workload improvements. In addition, pyEnGNet was used in a real-life study of patients after allogeneic stem cell transplantation with invasive aspergillosis and was able to detect biological perspectives in the study. [ABSTRACT FROM AUTHOR]

Because of the popularity of electric vehicles, large-scale charging piles are connected to the distribution network, so it is necessary to build an online platform for monitoring charging pile operation safety. In this paper, an online platform for monitoring charging pile operation safety was constructed from three aspects: hardware, database, and software functions using big data and related technologies. Firstly, the hardware platform was optimized by modifying the microcontroller and charging pile sensor equipment, adjusting the connection mode of the data acquisition circuit, and optimizing the topology of the communication network. Secondly, the mathematical model of the charging pile was built with the support of hardware equipment and a database, and the online operation data of the charging pile were collected by using big data technology and monitoring indicators such as voltage deviation and electric quantity deviation was selected to calculate the specific values of monitoring indicators. Finally, by comparing with the normal data of charging piles, the visual online monitoring results of charging piles were obtained, and it is concluded that the platform can provide decision support for the safe operation of charging piles. [ABSTRACT FROM AUTHOR]