Search results

This study aims to develop a machine learning model to detect structure fire fatalities using a dataset comprising 11,341 cases from 2011 to 2019.

Exploratory data analysis (EDA) was conducted prior to modelling, in which ten machine learning models were experimented with.

The main fatal structure fire risk factors were fires originating from bedrooms, living areas and the cooking/dining areas. The highest fatality rate (20.69%) was reported for fires ignited due to bedding (23.43%), despite a low fire incident rate (3.50%). Using 21 structure fire features, Random Forest (RF) yielded the best detection performance with 86% accuracy, followed by Decision Tree (DT) with bagging (accuracy = 84.7%).

Limitations of the study are pertaining to data quality and grouping of categories in the data pre-processing stage, which could affect the performance of the models.

The study is the first of its kind to manipulate risk factors to detect fatal structure classification, particularly focussing on structure fire fatalities. Most of the previous studies examined the importance of fire risk factors and their relationship to the fire risk level.

Purpose: The COVID-19 pandemic has been a good example of a Volatility, Uncertainty, Complexity, and Ambiguity (VUCA) world. Higher educational institutions (HEIs) have faced a massive hit because the jobs in this industry have become unexpected. Considering the most valuable assets ‘Teachers’ crunched in the VUCA crisis, the study intends to determine if personal harmony (PH) and organisational citizenship behaviour (OCB) would enhance teachers’ job satisfaction (JS).

Design/methodology/approach: Data are collected from the teachers of Indian HEIs and teachers who have experienced the impact of the COVID-19 catastrophe (VUCA). Considering the pandemic restrictions, data have been collected through an online survey (N = 364).

Practical Implications: PH is an individual’s internal quality and attribute that cannot be developed on force or situational need. Even in an uncertain situation, teachers have tried their best to contribute through professional service. Hence, people who possess PH contribute their best even though unsatisfied with their jobs.

Originality/value: This study has focused on finding the relationship between two different variables, PH and OCB (which has not been explored in Asian countries, majorly in India, where it has a vast cultural diversity and structure influencing the educational policies) that hinders the factors influencing JS, where these two variables are highly influenced by hygiene factors such as values, culture, ethical standards, personal belief, leadership styles, and fair treatment showcased by the organisations/institutions.

This study aims to propose novel edge device-tailored federated learning algorithms of local classifiers (stochastic gradient descent, support vector machines), namely, FL-lSGD and FL-lSVM. These algorithms are designed to address the challenge of large-scale ImageNet classification.

The authors’ FL-lSGD and FL-lSVM trains in a parallel and incremental manner to build an ensemble local classifier on Raspberry Pis without requiring data exchange. The algorithms load small data blocks of the local training subset stored on the Raspberry Pi sequentially to train the local classifiers. The data block is split into k partitions using the k-means algorithm, and models are trained in parallel on each data partition to enable local data classification.

Empirical test results on the ImageNet data set show that the authors’ FL-lSGD and FL-lSVM algorithms with 4 Raspberry Pis (Quad core Cortex-A72, ARM v8, 64-bit SoC @ 1.5GHz, 4GB RAM) are faster than the state-of-the-art LIBLINEAR algorithm run on a PC (Intel(R) Core i7-4790 CPU, 3.6 GHz, 4 cores, 32GB RAM).

Efficiently addressing the challenge of large-scale ImageNet classification, the authors’ novel federated learning algorithms of local classifiers have been tailored to work on the Raspberry Pi. These algorithms can handle 1,281,167 images and 1,000 classes effectively.

The target of this research is to develop a mathematical model which combines the Resource-Constrained Multi-Project Scheduling Problem (RCMPSP) and the Multi-Skilled Resource-Constrained Project Scheduling Problem (MSRCPSP). Due to the importance of resource management, the proposed formulation comprises resource leveling considerations as well. The model aims to simultaneously optimize: (1) the total time to accomplish all projects and (2) the total deviation of resource consumptions from the uniform utilization levels.

The K-Means (KM) and Fuzzy C-Means (FCM) clustering methods have been separately applied to discover the clusters of activities which have the most similar resource demands. The discovered clusters are given to the scheduling process as priori knowledge. Consequently, the execution times of the activities with the most common resource requests will not overlap. The intricacy of the problem led us to incorporate the KM and FCM techniques into a meta-heuristic called the Bi-objective Symbiosis Organisms Search (BSOS) algorithm so that the real-life samples of this problem could be solved. Therefore, two clustering-based algorithms, namely, the BSOS-KM and BSOS-FCM have been developed.

Comparisons between the BSOS-KM, BSOS-FCM and the BSOS method without any clustering approach show that the clustering techniques could enhance the optimization process. Another hybrid clustering-based methodology called the NSGA-II-SPE has been added to the comparisons to evaluate the developed resource leveling framework.

The practical importance of the model and the clustering-based algorithms have been demonstrated in planning several construction projects, where multiple water supply systems are concurrently constructed.

Reviewing the literature revealed that there was a need for a hybrid formulation that embraces the characteristics of the RCMPSP and MSRCPSP with resource leveling considerations. Moreover, the application of clustering algorithms as resource leveling techniques was not studied sufficiently in the literature.

With the development of inclusive financial business in China in recent years, this case describes the credit risk control of “mobile credit”, a smart online credit platform launched by Shanghai Mobanker Co. Ltd. (referred to as “Mobanker”, previously named as “Shanghai Mobanker Financial Information Service Co., Ltd.”) which provides technical services for inclusive finance industry.

Lack of appropriate student support and drawbacks in academic progression signify the importance of enhancing assessment and feedback in higher education (HE). Although assessment and feedback are significant in HE, minimal empirical research holistically explores the best practices. This study aims to address the niche and develop a decisive guideline for enhancing assessment setting and feedback provision within HE curricula.

A systematic approach was taken to obtain data for the study: a literature review underpinning the thematic content analysis of study documents, followed by semi-structured interviews. Document analysis contained mid-module reviews/student feedback; rubrics used in assessment; and formative/summative feedback provided for the graded work. Documental analysis informed the key attributes of the semi-structured interview. Interpretive structural modelling (ISM) analysis identified the influence and reliance of each driver.

This study revealed 15 drivers – 4 fundamental, 6 significant and 5 important – for enhancing assessment and feedback. The level partitioning from the ISM analysis established that all assessment and feedback needs to be underpinned by the university policy and fed into the assessment regime and marking scheme. This study identified that National Student Survey results were significantly improved due to implementing said drivers compared with the national and sector benchmarks.

The developed drivers enable the best practices in assessment setting and feedback provision. The level partition diagram can be used as a decisive guideline or a provisional framework in assessment and feedback provision for quality assurance in HE.

This study is one of, if not the only, to develop a guideline for signposting drivers and their influence and reliance to enhance assessment and feedback in a holistic HE setting. The developed drivers and the level partition diagram bring novelty and add to the current body of knowledge.

Spam emails classification using data mining and machine learning approaches has enticed the researchers' attention duo to its obvious positive impact in protecting internet users. Several features can be used for creating data mining and machine learning based spam classification models. Yet, spammers know that the longer they will use the same set of features for tricking email users the more probably the anti-spam parties might develop tools for combating this kind of annoying email messages. Spammers, so, adapt by continuously reforming the group of features utilized for composing spam emails. For that reason, even though traditional classification methods possess sound classification results, they were ineffective for lifelong classification of spam emails duo to the fact that they might be prone to the so-called “Concept Drift”. In the current study, an enhanced model is proposed for ensuring lifelong spam classification model. For the evaluation purposes, the overall performance of the suggested model is contrasted against various other stream mining classification techniques. The results proved the success of the suggested model as a lifelong spam emails classification method.

This study aims to recognize the significant barriers of small and medium-sized enterprises (SMEs) in Bangladesh, hindering the adoption of total quality management (TQM). Additionally, this research intends to explore the interrelations among these barriers to develop essential managerial insights for promoting TQM implementation in SMEs.

The interpretive structural modeling (ISM) approach and Matrice d'impacts croisés multiplication appliquée á un classment (MICMAC) a cross-impact matrix multiplication applied to classification show the relationship among the barriers and classification of the barriers to TQM implementation respectively, and partial least squares structural equation modeling (PLS-SEM) is applied for ISM model validation.

This study examined previous literature and conducted interviews with professionals to identify 17 barriers. The study then develops and investigates a model that outlines the relationships and priorities among these barriers and categorizes them based on their impact and interdependence. This analysis can assist SMEs in implementing TQM during their operations successfully.

This research emphasizes the crucial obstacles that greatly affect other barriers and require immediate attention. Furthermore, this study provides valuable information for SMEs to effectively prioritize their resources and efforts to overcome these obstacles.

This study delves into the primary obstacles impeding the integration of TQM in SMEs through a novel approach. Additionally, this study constructs a verified contextual framework that depicts the hierarchies and interconnections among these barriers.

This study focuses on the classification of targets with varying shapes using radar cross section (RCS), which is influenced by the target’s shape. This study aims to develop a robust classification method by considering an incident angle with minor random fluctuations and using a physical optics simulation to generate data sets.

The approach involves several supervised machine learning and classification methods, including traditional algorithms and a deep neural network classifier. It uses histogram-based definitions of the RCS for feature extraction, with an emphasis on resilience against noise in the RCS data. Data enrichment techniques are incorporated, including the use of noise-impacted histogram data sets.

The classification algorithms are extensively evaluated, highlighting their efficacy in feature extraction from RCS histograms. Among the studied algorithms, the K-nearest neighbour is found to be the most accurate of the traditional methods, but it is surpassed in accuracy by a deep learning network classifier. The results demonstrate the robustness of the feature extraction from the RCS histograms, motivated by mm-wave radar applications.

This study presents a novel approach to target classification that extends beyond traditional methods by integrating deep neural networks and focusing on histogram-based methodologies. It also incorporates data enrichment techniques to enhance the analysis, providing a comprehensive perspective for target detection using RCS.

This work aims to establish the relationship between painting art and sustainability, which allows for highlighting implications likely to improve sustainability for humanity's welfare.

To achieve this objective, painting art is measured by a composite index aggregating the quantity and quality represented by the market value. As for sustainable development, it is represented by a composite index comprising three variables: the climate change performance index (ecological dimension), the wage index reflecting distributive justice (social dimension) and the gross domestic product (economic dimension). The composite indices were determined through adjusted data envelopment analysis. In addition, two other methods are used in this work: correlation analysis and a neural network method. These methods are applied to data from 2007 to 2021 across the world.

The correlation method highlighted a perfect positive correlation between painting art and sustainability. As for the neural network method, it revealed that the quality of painting has the greatest impact on sustainability. The neural network method also showed that the most positively impacted variable of sustainability by painting art is the social variable, with a pseudo-probability of 0.90.

The relationship between painting art and sustainability is underexplored, in particular in terms of statistical analysis. Therefore, this research intends to fill this gap. Moreover, analysis of the relationship between both using composite indices computed via an original method (adjusted data envelopment analysis) and a neural network method is nonexistent, which constitutes the novelty of this work.

The peer review history for this article is available at: https://publons.com/publon/10.1108/IJSE-01-2023-0006