Search results

Surface roughness has a serious impact on the fatigue strength, wear resistance and life of mechanical products. Realizing the evolution of surface quality through theoretical modeling takes a lot of effort. To predict the surface roughness of milling processing, this paper aims to construct a neural network based on deep learning and data augmentation.

This study proposes a method consisting of three steps. Firstly, the machine tool multisource data acquisition platform is established, which combines sensor monitoring with machine tool communication to collect processing signals. Secondly, the feature parameters are extracted to reduce the interference and improve the model generalization ability. Thirdly, for different expectations, the parameters of the deep belief network (DBN) model are optimized by the tent-SSA algorithm to achieve more accurate roughness classification and regression prediction.

The adaptive synthetic sampling (ADASYN) algorithm can improve the classification prediction accuracy of DBN from 80.67% to 94.23%. After the DBN parameters were optimized by Tent-SSA, the roughness prediction accuracy was significantly improved. For the classification model, the prediction accuracy is improved by 5.77% based on ADASYN optimization. For regression models, different objective functions can be set according to production requirements, such as root-mean-square error (RMSE) or MaxAE, and the error is reduced by more than 40% compared to the original model.

A roughness prediction model based on multiple monitoring signals is proposed, which reduces the dependence on the acquisition of environmental variables and enhances the model's applicability. Furthermore, with the ADASYN algorithm, the Tent-SSA intelligent optimization algorithm is introduced to optimize the hyperparameters of the DBN model and improve the optimization performance.

Research and development (R&D) projects are often pursued through a project portfolio (PP). R&D PPs involve many stakeholders, and without proactive management, their interactions may lead to conflict risks. These conflict risks change dynamically with different stages of the PP life cycle, increasing the challenge of PP risk management. Existing conflict risk research mainly focuses on source identification but lacks risk assessment work. To better manage the stakeholder conflict risks (SCRs) of R&D PPs, this study employs the dynamic Bayesian network (DBN) to construct its dynamic assessment model.

This study constructs a DBN model to assess the SCRs in R&D PP. First, an indicator system of SCRs is constructed from the life cycle perspective. Then, the risk relationships within each R&D PPs life cycle stage are identified via interpretative structural modeling (ISM). The prior and conditional probabilities of risks are obtained by expert judgment and Monte Carlo simulation (MCS). Finally, crucial SCRs at each stage are identified utilizing propagation analysis, and the corresponding risk responses are proposed.

The results of the study identify the crucial risks at each stage. Also, for the crucial risks, this study suggests appropriate risk response strategies to help managers better perform risk response activities.

This study dynamically assesses the stakeholder conflict risks in R&D PPs from a life-cycle perspective, extending the stakeholder risk management research. Meanwhile, the crucial risks are identified at each stage accordingly, providing managerial insights for R&D PPs.

The energy efficiency of buildings has been emphasized along with the continual development in the building and construction sector that consumes a significant amount of energy. To this end, the purpose of this research paper is to forecast energy consumption to improve energy resource planning and management.

This study proposes the application of the convolutional neural network (CNN) for estimating the electricity consumption in the Grey Nuns building in Canada. The performance of the proposed model is compared against that of long short-term memory (LSTM) and multilayer perceptron (MLP) neural networks. The models are trained and tested using monthly electricity consumption records (i.e. from May 2009 to December 2021) available from Concordia’s facility department. Statistical measures (e.g. determination coefficient [R²], root mean squared error [RMSE], mean absolute error [MAE] and mean absolute percentage error [MAPE]) are used to evaluate the outcomes of models.

The results reveal that the CNN model outperforms the other model predictions for 6 and 12 months ahead. It enhances the performance metrics reported by the LSTM and MLP models concerning the R², RMSE, MAE and MAPE by more than 4%, 6%, 42% and 46%, respectively. Therefore, the proposed model uses the available data to predict the electricity consumption for 6 and 12 months ahead. In June and December 2022, the overall electricity consumption is estimated to be 195,312 kWh and 254,737 kWh, respectively.

This study discusses the development of an effective time-series model that can forecast future electricity consumption in a Canadian heritage building. Deep learning techniques are being used for the first time to anticipate the electricity consumption of the Grey Nuns building in Canada. Additionally, it evaluates the effectiveness of deep learning and machine learning methods for predicting electricity consumption using established performance indicators. Recognizing electricity consumption in buildings is beneficial for utility providers, facility managers and end users by improving energy and environmental efficiency.

Deep learning (DL) is on the rise because it can make predictions and judgments based on data that is unseen. Blockchain technologies are being combined with DL frameworks in various industries to provide a safe and effective infrastructure. The review comprises literature that lists the most recent techniques used in the aforementioned application sectors. We examine the current research trends across several fields and evaluate the literature in terms of its advantages and disadvantages.

The integration of blockchain and DL has been explored in several application domains for the past five years (2018–2023). Our research is guided by five research questions, and based on these questions, we concentrate on key application domains such as the usage of Internet of Things (IoT) in several applications, healthcare and cryptocurrency price prediction. We have analyzed the main challenges and possibilities concerning blockchain technologies. We have discussed the methodologies used in the pertinent publications in these areas and contrasted the research trends during the previous five years. Additionally, we provide a comparison of the widely used blockchain frameworks that are used to create blockchain-based DL frameworks.

By responding to five research objectives, the study highlights and assesses the effectiveness of already published works using blockchain and DL. Our findings indicate that IoT applications, such as their use in smart cities and cars, healthcare and cryptocurrency, are the key areas of research. The primary focus of current research is the enhancement of existing systems, with data analysis, storage and sharing via decentralized systems being the main motivation for this integration. Amongst the various frameworks employed, Ethereum and Hyperledger are popular among researchers in the domain of IoT and healthcare, whereas Bitcoin is popular for research on cryptocurrency.

There is a lack of literature that summarizes the state-of-the-art methods incorporating blockchain and DL in popular domains such as healthcare, IoT and cryptocurrency price prediction. We analyze the existing research done in the past five years (2018–2023) to review the issues and emerging trends.

By introducing Stacking algorithm to solve the underfitting problem caused by insufficient data in traditional machine learning, this paper provides a new solution to the cold start problem of entrepreneurial borrowing risk control.

The authors introduce semi-supervised learning and integrated learning into the field of migration learning, and innovatively propose the Stacking model migration learning, which can independently train models on entrepreneurial borrowing credit data, and then use the migration strategy itself as the learning object, and use the Stacking algorithm to combine the prediction results of the source domain model and the target domain model.

The effectiveness of the two migration learning models is evaluated with real data from an entrepreneurial borrowing. The algorithmic performance of the Stacking-based model migration learning is further improved compared to the benchmark model without migration learning techniques, with the model area under curve value rising to 0.8. Comparing the two migration learning models reveals that the model-based migration learning approach performs better. The reason for this is that the sample-based migration learning approach only eliminates the noisy samples that are relatively less similar to the entrepreneurial borrowing data. However, the calculation of similarity and the weighing of similarity are subjective, and there is no unified judgment standard and operation method, so there is no guarantee that the retained traditional credit samples have the same sample distribution and feature structure as the entrepreneurial borrowing data.

From a practical standpoint, on the one hand, it provides a new solution to the cold start problem of entrepreneurial borrowing risk control. The small number of labeled high-quality samples cannot support the learning and deployment of big data risk control models, which is the cold start problem of the entrepreneurial borrowing risk control system. By extending the training sample set with auxiliary domain data through suitable migration learning methods, the prediction performance of the model can be improved to a certain extent and more generalized laws can be learned.

This paper introduces the thought method of migration learning to the entrepreneurial borrowing scenario, provides a new solution to the cold start problem of the entrepreneurial borrowing risk control system and verifies the feasibility and effectiveness of the migration learning method applied in the risk control field through empirical data.

The aim of this study is to investigate subject independent emotion recognition capabilities of EEG and peripheral physiological signals namely: electroocoulogram (EOG), electromyography (EMG), electrodermal activity (EDA), temperature, plethysmograph and respiration. The experiments are conducted on both modalities independently and in combination. This study arranges the physiological signals in order based on the prediction accuracy obtained on test data using time and frequency domain features.

DEAP dataset is used in this experiment. Time and frequency domain features of EEG and physiological signals are extracted, followed by correlation-based feature selection. Classifiers namely – Naïve Bayes, logistic regression, linear discriminant analysis, quadratic discriminant analysis, logit boost and stacking are trained on the selected features. Based on the performance of the classifiers on the test set, the best modality for each dimension of emotion is identified.

 The experimental results with EEG as one modality and all physiological signals as another modality indicate that EEG signals are better at arousal prediction compared to physiological signals by 7.18%, while physiological signals are better at valence prediction compared to EEG signals by 3.51%. The valence prediction accuracy of EOG is superior to zygomaticus electromyography (zEMG) and EDA by 1.75% at the cost of higher number of electrodes. This paper concludes that valence can be measured from the eyes (EOG) while arousal can be measured from the changes in blood volume (plethysmograph). The sorted order of physiological signals based on arousal prediction accuracy is plethysmograph, EOG (hEOG + vEOG), vEOG, hEOG, zEMG, tEMG, temperature, EMG (tEMG + zEMG), respiration, EDA, while based on valence prediction accuracy the sorted order is EOG (hEOG + vEOG), EDA, zEMG, hEOG, respiration, tEMG, vEOG, EMG (tEMG + zEMG), temperature and plethysmograph.

Many of the emotion recognition studies in literature are subject dependent and the limited subject independent emotion recognition studies in the literature report an average of leave one subject out (LOSO) validation result as accuracy. The work reported in this paper sets the baseline for subject independent emotion recognition using DEAP dataset by clearly specifying the subjects used in training and test set. In addition, this work specifies the cut-off score used to classify the scale as low or high in arousal and valence dimensions. Generally, statistical features are used for emotion recognition using physiological signals as a modality, whereas in this work, time and frequency domain features of physiological signals and EEG are used. This paper concludes that valence can be identified from EOG while arousal can be predicted from plethysmograph.

Since conducting agile strategies provides sustainable passenger satisfaction and revenue by replacing applied policies with more profitable ones rapidly, the focus of this study is to evaluate agile attributes for managing low-cost carriers (LCCs) operations by means of resources and competences based on dynamic capabilities built on resource-based view (RBV) theory and to achieve sustainable competitive advantage in a volatile and dynamic air transport environment. LCCs in Turkey are also evaluated in this study since the competition among LCCs is high to gain market share and they can adapt quickly to all kinds of circumstances.

Two well-known Multi-Criteria Decision-Making Methods (MCDM) named as the Stepwise Weight Assessment Ratio Analysis (SWARA) and multi-attributive border approximation area comparison (MABAC) methods by employing Picture fuzzy sets (PiFS) are employed to determine weight of agile attributes and superiority of LCCs based on agile attributes in the market, respectively. To check the consistency and robustness of the results for the proposed approach, comparative and sensitivity analysis are performed at the end of the study.

While the ranking orders of agile attributes are Strategic Responsiveness (AG1), Financial Management (AG4), Quality (AG2), Digital integration (AG3) and Reliability (AG5), respectively, LCC2 is selected as the best agile airline company in Turkey with respect to agile attributes. SWARA and MABAC method based on PiFS is appropriate and effective method to evaluate agile attributes that has important reference value for the airline companies in aviation industry.

The findings of this study will support managers in the airline industry to conduct airline operations more flexibly and effectively to take sustainable competitive advantage in unexpected and dynamic environment.

To the author' best knowledge, this study is the first developed to identify the attributes necessary to increase agility in LCCs. Thus, as a systematic tool, a framework is developed for the implementation of agile attributes to achieve sustainable competitive advantage in the airline industry and presented a roadmap for airline managers to deal with crises and challenging situations by satisfying customer and increasing competitiveness.

Machine learning (ML), and deep learning in particular, is gaining traction across a myriad of real-life applications. Portfolio management is no exception. This paper provides a systematic literature review of deep learning applications for portfolio management. The findings are likely to be valuable for industry practitioners and researchers alike, experimenting with novel portfolio management approaches and furthering investment management practice.

This review follows the guidance and methodology of Linnenluecke et al. (2020), Massaro et al. (2016) and Fisch and Block (2018) to first identify relevant literature based on an appropriately developed search phrase, filter the resultant set of publications and present descriptive and analytical findings of the research itself and its metadata.

The authors find a strong dominance of reinforcement learning algorithms applied to the field, given their through-time portfolio management capabilities. Other well-known deep learning models, such as convolutional neural network (CNN) and recurrent neural network (RNN) and its derivatives, have shown to be well-suited for time-series forecasting. Most recently, the number of papers published in the field has been increasing, potentially driven by computational advances, hardware accessibility and data availability. The review shows several promising applications and identifies future research opportunities, including better balance on the risk-reward spectrum, novel ways to reduce data dimensionality and pre-process the inputs, stronger focus on direct weights generation, novel deep learning architectures and consistent data choices.

Several systematic reviews have been conducted with a broader focus of ML applications in finance. However, to the best of the authors’ knowledge, this is the first review to focus on deep learning architectures and their applications in the investment portfolio management problem. The review also presents a novel universal taxonomy of models used.

This paper undertakes an extensive and systematic review of the literature on earnings management (EM) over the past three decades (1992–2022). Furthermore, the study identifies emerging research themes and proposes future avenues for further investigation in the realm of EM.

For this study, a comprehensive collection of 2,775 articles on EM published between 1992 and 2022 was extracted from the Scopus database. The author employed various tools, including Microsoft Excel, R studio, Gephi and visualization of similarities viewer, to conduct bibliometric, content, thematic and cluster analyses. Additionally, the study examined the literature across three distinct periods: prior to the enactment of the Sarbanes-Oxley Act (1992–2001), subsequent to the implementation of the Sarbanes-Oxley Act (2002–2012), and after the adoption of International Financial Reporting Standards (2013–2022) to draw more inferences and insights on EM research.

The study identifies three major themes, namely the operationalization of EM constructs, the trade-off between EM tools (accrual EM, real EM and classification shifting) and the role of corporate governance in mitigating EM in emerging markets. Existing literature in these areas presents mixed and inconclusive findings, suggesting the need for further theoretical development. Further, the study findings observe a shift in research focus over time: initially, understanding manipulation techniques, then evaluating regulatory measures, and more recently, investigating the impact of global accounting standards. Several emerging research themes (technology advancements, cross-cultural and cross-national studies, sustainability, behavioral aspects and non-financial indicators of EM) have been identified. This study subsequent analysis reveals an evolving EM landscape, with researchers from disciplines like data science, computer science and engineering applying their analytical expertise to detect EM anomalies. Furthermore, this study offers significant insights into sophisticated EM techniques such as neural networks, machine learning techniques and hidden Markov models, among others, as well as relevant theories including dynamic capabilities theory, learning curve theory, psychological contract theory and normative institutional theory. These techniques and theories demonstrate the need for further advancement in the field of EM. Lastly, the findings shed light on prominent EM journals, authors and countries.

This study conducts quantitative bibliometric and thematic analyses of the existing literature on EM while identifying areas that require further development to advance EM research.

Information technology (IT) personnels’ technical, business and behavioral skills are critical enablers for generating IT value. In an increasingly digitalized working environment where non-IT employees participate in digital innovations, a focus on IT personnels’ skills only doesn’t meet researchers’ need for a framework to study digital skills and managers’ need to address digital skills challenges across an enterprise’s workforce. Nevertheless, the digital skills topic is complicated by conceptual ambiguity and a lack of theoretically derived and empirically validated model. The purpose of this study is to address this problem.

Theoretically, this study draws on human capital (HC) and resource-based view (RBV) theories. Empirically, it follows mixed method combining interviews and a survey.

The digital skills construct is a multidimensional second order reflective construct. While its development is influenced by an organization’s commitment and exposure to digitalization, it influences the value organizations obtain from digitalization.

This study conceptualizes the digital skills construct, identifying technology agnostic subdimensions that are meaningful beyond a particular digital domain [information and communication technology (ICT), information, Internet, Inter of Things (IoT)] and establishing a valid measure. Other researchers can improve both the indicators of the existing four conceptually distinct and managerially recognizable workplace digital skills dimensions as well as testing new ones.

Managers can use the instrument to assess the extent to which their non-IT workforces are equipped with digital skills and get strategic insights for specific interventions such as upskilling or buying in skills.

The main theoretical contribution of the paper is the conceptualization and validation of the digital skills construct for the non-IT workforce. Furthermore, we provide a theoretical framework to explain the factors that could influence the development of digital skills and demonstrate the impact that digital skills have on selected digitalization value indicators. This contribution provides the foundation for investigating the drivers, outcomes and the relationship of digital skills to other constructs such as digital transformation, innovation and firm performance.