Search results

Software defect prediction (SDP) is a critical aspect of software quality assurance, aiming to identify and manage potential defects in software systems. In this paper, we have proposed a novel hybrid approach that combines Gray Wolf Optimization with Feature Selection (GWOFS) and multilayer perceptron (MLP) for SDP. The GWOFS-MLP hybrid model is designed to optimize feature selection, ultimately enhancing the accuracy and efficiency of SDP. Gray Wolf Optimization, inspired by the social hierarchy and hunting behavior of gray wolves, is employed to select a subset of relevant features from an extensive pool of potential predictors. This study investigates the key challenges that traditional SDP approaches encounter and proposes promising solutions to overcome time complexity and the curse of the dimensionality reduction problem.

The integration of GWOFS and MLP results in a robust hybrid model that can adapt to diverse software datasets. This feature selection process harnesses the cooperative hunting behavior of wolves, allowing for the exploration of critical feature combinations. The selected features are then fed into an MLP, a powerful artificial neural network (ANN) known for its capability to learn intricate patterns within software metrics. MLP serves as the predictive engine, utilizing the curated feature set to model and classify software defects accurately.

The performance evaluation of the GWOFS-MLP hybrid model on a real-world software defect dataset demonstrates its effectiveness. The model achieves a remarkable training accuracy of 97.69% and a testing accuracy of 97.99%. Additionally, the receiver operating characteristic area under the curve (ROC-AUC) score of 0.89 highlights the model’s ability to discriminate between defective and defect-free software components.

Experimental implementations using machine learning-based techniques with feature reduction are conducted to validate the proposed solutions. The goal is to enhance SDP’s accuracy, relevance and efficiency, ultimately improving software quality assurance processes. The confusion matrix further illustrates the model’s performance, with only a small number of false positives and false negatives.

The traditional failure mode and effect analysis (FMEA) has some limitations, such as the neglect of relevant historical data, subjective use of rating numbering and the less rationality and accuracy of the Risk Priority Number. The current study proposes a machine learning–enhanced FMEA (ML-FMEA) method based on a popular machine learning tool, Waikato environment for knowledge analysis (WEKA).

This work uses the collected FMEA historical data to predict the probability of component/product failure risk by machine learning based on different commonly used classifiers. To compare the correct classification rate of ML-FMEA based on different classifiers, the 10-fold cross-validation is employed. Moreover, the prediction error is estimated by repeated experiments with different random seeds under varying initialization settings. Finally, the case of the submersible pump in Bhattacharjee et al. (2020) is utilized to test the performance of the proposed method.

The results show that ML-FMEA, based on most of the commonly used classifiers, outperforms the Bhattacharjee model. For example, the ML-FMEA based on Random Committee improves the correct classification rate from 77.47 to 90.09 per cent and area under the curve of receiver operating characteristic curve (ROC) from 80.9 to 91.8 per cent, respectively.

The proposed method not only enables the decision-maker to use the historical failure data and predict the probability of the risk of failure but also may pave a new way for the application of machine learning techniques in FMEA.

Warranty-based big data analysis has attracted a great deal of attention because of its key capabilities and role in improving product quality while minimizing costs. Information and details about particular parts (components) repair and replacement during the warranty term, usually stored in the after-sales service database, can be used to solve problems in a variety of sectors. Due to the small number of studies related to the complete analysis of parts failure patterns in the automotive industry in the literature, this paper focuses on discovering and assessing the impact of lesser-studied factors on the failure of auto parts in the warranty period from the after-sales data of an automotive manufacturer.

The interconnected method used in this study for analyzing failure patterns is formed by combining association rules (AR) mining and Bayesian networks (BNs).

This research utilized AR analysis to extract valuable information from warranty data, exploring the relationship between component failure, time and location. Additionally, BNs were employed to investigate other potential factors influencing component failure, which could not be identified using Association Rules alone. This approach provided a more comprehensive evaluation of the data and valuable insights for decision-making in relevant industries.

This study's findings are believed to be practical in achieving a better dissection and providing a comprehensive package that can be utilized to increase component quality and overcome cross-sectional solutions. The integration of these methods allowed for a wider exploration of potential factors influencing component failure, enhancing the validity and depth of the research findings.

The objective of the proposed work is to identify the most commonly occurring non–small cell carcinoma types, such as adenocarcinoma and squamous cell carcinoma, within the human population. Another objective of the work is to reduce the false positive rate during the classification.

In this work, a hybrid method using convolutional neural networks (CNNs), extreme gradient boosting (XGBoost) and long-short-term memory networks (LSTMs) has been proposed to distinguish between lung adenocarcinoma and squamous cell carcinoma. To extract features from non–small cell lung carcinoma images, a three-layer convolution and three-layer max-pooling-based CNN is used. A few important features have been selected from the extracted features using the XGBoost algorithm as the optimal feature. Finally, LSTM has been used for the classification of carcinoma types. The accuracy of the proposed method is 99.57 per cent, and the false positive rate is 0.427 per cent.

The proposed CNN–XGBoost–LSTM hybrid method has significantly improved the results in distinguishing between adenocarcinoma and squamous cell carcinoma. The importance of the method can be outlined as follows: It has a very low false positive rate of 0.427 per cent. It has very high accuracy, i.e. 99.57 per cent. CNN-based features are providing accurate results in classifying lung carcinoma. It has the potential to serve as an assisting aid for doctors.

It can be used by doctors as a secondary tool for the analysis of non–small cell lung cancers.

It can help rural doctors by sending the patients to specialized doctors for more analysis of lung cancer.

In this work, a hybrid method using CNN, XGBoost and LSTM has been proposed to distinguish between lung adenocarcinoma and squamous cell carcinoma. A three-layer convolution and three-layer max-pooling-based CNN is used to extract features from the non–small cell lung carcinoma images. A few important features have been selected from the extracted features using the XGBoost algorithm as the optimal feature. Finally, LSTM has been used for the classification of carcinoma types.

Color psychology theory reveals that complex images with very varied palettes and many different colors are likely to be considered unattractive by individuals. Notwithstanding, web content containing social signals may be more attractive via the initiation of a social connection. This research investigates a predictive model blending variables from these theoretical perspectives to determine crowdfunding success.

The research is based on data from 176,614 Kickstarter projects. A number of machine learning and artificial intelligence techniques were employed to analyze the listing images for color complexity and the presence of people, while specific language features, including socialness, were measured in the listing text. Logistic regression was applied, controlling for several additional variables and predictive model was developed.

The findings supported the color complexity and socialness effects on crowdfunding success. The model achieves notable predictive value explaining 56.4% of variance. Listing images containing fewer colors and that have more similar colors are more likely to be crowdfunded successfully. Listings that convey greater socialness have a greater likelihood of being funded.

This investigation contributes a unique understanding of the effect of features of both socialness and color complexity on the success of crowdfunding ventures. A second contribution comes from the process and methods employed in the study, which provides a clear blueprint for the processing of large-scale analysis of soft information (images and text) in order to use them as variables in the scientific testing of theory.

This paper aims first to describe the most prevalent teachers’ and students’ behaviors in synchronous online classes in emergency remote teaching; second, to discern behavior profiles and third, to investigate what features explain the observed behaviors.

An adapted COPUS observation protocol was employed to observe 292 online classes from 146 higher education teachers.

The most prevalent behaviors were: Presenting for teachers and Receiving for students, followed by Teachers Guiding and Students Talking to Class. Furthermore, cluster analysis showed two groups: Traditional and Interactive. The variables that better explained belonging to the Interactive lecture group were disciplinary area – social sciences and humanities –and teaching in technical institutions.

In a context where higher education institutions intend to project the lessons learned into post-pandemic learning experiences, this study provides observational evidence to realize the full potential expected from online and blended teaching and learning.

Despite the prevalence of synchronous online lectures during COVID-19, there is a paucity of observational studies on the actual behaviors that occurred in this context. Most research has been based on surveys and interviews. This study addresses this gap.

Wine tourism is gaining increasing popularity among Chinese tourists, making it necessary to thoroughly examine tourist behavior. While online reviews posted by wine tourists have been extensively studied from the perspectives of destinations and wineries, the perspective of the tourists themselves has been overlooked. To address this gap, this study aims to identify significant attributes intrinsic to the tourism experiences of Chinese wine tourists by adopting a text-mining approach from a tourist-centric perspective.

The authors use topic modeling to extract these attributes, calculate topic intensity to understand tourists’ attention distribution across these attributes and conduct topical sentiment analysis to evaluate tourists’ satisfaction levels with each attribute. The authors perform importance-performance analyses (IPAs) using topic intensity and sentiment scores. Furthermore, the authors conduct semistructured in-depth interviews with Chinese wine tourists to gain insights into the underlying reasons behind the key findings.

The study identifies eleven attributes for domestic wine tourists and seven attributes for outbound wine tourists. From the reviews of both domestic and outbound tourists, three common attributes have been identified: “scenic view”, “wine tasting and purchase” and “wine knowledge”.

According to the results of the IPAs, there is a pressing need for enhancements in the wine tasting and purchasing experience at domestic wine attractions. Additionally, managers of domestic wine attractions should continue to prioritize the positive aspects of the family trip experience and scenic views. On the other hand, for outbound wine attractions, it is crucial for managers to maintain their efforts in providing opportunities for wine knowledge acquisition, ensuring scenic views and upholding the reputation of wine regions.

First, this study breaks new ground by adopting a tourist-centric perspective to extract significant attributes from real wine tourism reviews. Second, the authors conduct a comparative analysis between Chinese wine tourists who travel domestically and those who travel abroad. The third novel aspect of this study is the application of IPA based on textual review data in the context of wine tourism. Fourth, by integrating topic modeling with qualitative interviews, the authors use a mixed-method approach to gain deeper insights into the experiences of Chinese wine tourists.

Although green entrepreneurial orientation (GEO) has received much attention, it is unclear whether it affects technological green innovation (GI). Therefore, this study aims to understand how GEO affects technological GI, with its dimensions green product innovation (GPRODI) and green process innovation (GPROCI), as well as to explore whether resource orchestration capability (ROC) moderates the relationships between them.

Based on a cross-sectional survey design, data were gathered from 177 managers of large manufacturing firms in Yemen and analysed using partial least squares structural equation modelling via SmartPLS software.

The results revealed that GEO positively affects both GPRODI and GPROCI, with a higher effect on GPROCI. Importantly, ROC does, in fact, positively moderate the link between GEO and GPRODI.

This research adds to knowledge by combining GEO, ROC and technological GI into a unified framework, considering the perspectives of the resource-based view and the resource orchestration theory. However, the study’s use of cross-sectional survey data makes it impossible to infer causes. This is because GEO, ROC and technological GI all have effects on time that this empirical framework cannot account for.

The findings from this research provide valuable insights for executives and decision makers of large manufacturing companies, who are expected to show increasing interest in adopting ROC into their organisations. This suggests that environmentally-conscious entrepreneurial firms can enhance their GI efforts by embracing ROC.

By adopting the proposed framework, firms can carry out their activities in ways that do not harm environmental and societal well-being, as simply achieving high economic performance is no longer sufficient.

Theoretically, the results offer an in-depth understanding of the role of GEO in the technological GI domain by indicating that GEO can promote GPRODI and GPROCI. In addition, the results shed new light on the boundaries of GEO from the perspective of resource orchestration theory. Furthermore, the findings present important insights for managers aiming to enhance their comprehension of leveraging GEO and ROC to foster technological GI.

This research explores the link between job redesign and skilling in three Lebanese service provider industries, aiming to understand how these factors affect organizational agility, a crucial factor for efficiency and effectiveness and promote long-term interventions through job redesign, upskilling and reskilling.

This study employed two surveys, one for personnel (employees) and one for human aid managers (HR managers). These surveys collected data from 384 employees and 67 HR managers. The study utilized a work design questionnaire (WDQ), skilling application evaluation and the change acceptance model and testing to evaluate job redesign, skilling application effectiveness, technology acceptance and change readiness.

It was revealed that there is a significant and positive relationship between job redesign and the application of skilling programs. This relationship was shown to enhance organizational agility, with a particular focus on employees' technology acceptance and readiness for change. The integrated framework that combines job redesign, upskilling and reskilling was empirically tested and found to enable organizations to build their agility. The study also identified challenges and offered solutions for implementation, emphasizing the importance of employee responsiveness.

This research emphasizes the need for organizations to adapt job designs and enhance employee skills to enhance organizational agility, recommending a structured approach that combines job redesign and skill development efforts.

This research integrates job redesign, upskilling and reskilling in Lebanese service provider industries, contributing to organizational change and workforce development. It emphasizes technology acceptance and readiness for change.

This paper aims to provide a complete analysis framework and prediction method for the construction of the patent securitization (PS) basic asset pool.

This paper proposes an integrated classification method based on genetic algorithm and random forest algorithm. First, comprehensively consider the patent value evaluation model and SME credit evaluation model, determine 17 indicators to measure the patent value and SME credit; Secondly, establish the classification label of high-quality basic assets; Then, genetic algorithm and random forest model are used to predict and screen high-quality basic assets; Finally, the performance of the model is evaluated.

The machine learning model proposed in this study is mainly used to solve the screening problem of high-quality patents that constitute the underlying asset pool of PS. The empirical research shows that the integrated classification method based on genetic algorithm and random forest has good performance and prediction accuracy, and is superior to the single method that constitutes it.

The main contributions of the article are twofold: firstly, the machine learning model proposed in this article determines the standards for high-quality basic assets; Secondly, this article addresses the screening issue of basic assets in PS.