This study presents a review of literature on the usage of artificial neural networks (ANNs) architecture contribution method and structural equation modeling (SEM), and proposes a new selection process in the context of algorithm -based SEM-ANNs schemes. This study enriches academic literature by providing a review of all the main aspects of customization in ANNs and contribution methods in combination with SEM. Academic databases are examined for exhibition findings, yielding 253 papers published between 2016 and 2022. The retrieved papers are categorized according to inclusion criteria, and the final set of 73 articles are discussed based on two directions, namely, ‘Sector-based’ and ‘Algorithm-based’ as a new representation of taxonomy research. A state-of-the-art bibliographic analysis is presented. This review also identifies modern challenges and open issues in terms of multiple evaluation criteria, importance criteria, and data variations related to the selection of customizations in ANNs and contribution methods combined with SEM in different industrial cases. Several issues fall under multicriteria decision making for handling complexity problems in different ANNs and contribution methods. Thus, this study also presents a research proposal and recommends a solution based on a three-phase methodology for handling the selection and overcoming the identified issues, subsequently completing a strategic guideline solution.

Theoretical models have become increasingly complex, but the dual-phase structural equation modelling (SEM) and artificial neural network analysis can be used by scholars to unveil the causal interactions and nonlinear relationships between variables. However, not only a single open issue and challenge—but several of them—are encountered in the use of different multi-assessment types of measurement model to achieve the reliability and validity whilst implementing SEM, but the gaps have not been fully determined at present. The issues significantly impact the effectiveness process of selecting the most suitable method to assess the measurement model of SEM. Once the best sequence quality improvement is met, it then needs to present a recommendable solution. To this end, this study completes the literature by presenting a systematic review of all main advanced aspects of the SEM reliability and validity approaches. Firstly, the databases of ScienceDirect, IEEE Xplore, Web of Science and Scopus were checked for the retrospective studies. A total of 239 papers were gathered for the period covering 2016 to June 2021. Then, the obtained articles were filtered according to the predefined inclusion criteria. Sixty articles were ultimately selected and divided into three categories (single, hybrid and other types) to enable a new representation of the crossover taxonomy amongst 'SEM reliability and validity' and 'multi-assessment methods for structural model' for the first time. The three categories had been matched with the SEM processes, and each of the detailed models were defined to determine the sets of principal criteria of the entire selected SEM approaches. Consequently, this multi-field interdisciplinary review was used to expose the state-of-the-art challenges and open issues (i.e. multiple-evaluation criteria, importance criteria and data variation) related to the sets of SEM criteria necessitating a selection process for deriving the best SEM method. Each issue entailed a 'wherefore', and multi-criteria decision making was adopted to handle the complexity problems in the different cases. Thus, a new three-phase decision-making methodology was constructed. In the first phase, a decision matrix (DM) was identified for the SEM approach; the composition of the decision alternatives and identified criteria were derived from the academic literature. In the second phase, the development methodology was achieved on the basis of the integrated multi-criteria DM techniques. The analytic hierarchy process was used for the subjective weighting of the criteria within the constructed DM, whereas the vlsekriterijumska optimizcija i kaompromisno resenje technique was used for ranking and selecting the best SEM methods. In the third phase, an objective validation approach was adopted to validate the proposed methodology. The outcome of this novel approach is intended to guide decision makers and policymakers on the easy evaluation of their goals of selecting the most suitable computing methods and the improvement of the reliability and validity of SEM.

Topical treatments with structural equation modelling (SEM) and an artificial neural network (ANN), including a wide range of concepts, benefits, challenges and anxieties, have emerged in various fields and are becoming increasingly important. Although SEM can determine relationships amongst unobserved constructs (i.e. independent, mediator, moderator, control and dependent variables), it is insufficient for providing non-compensatory relationships amongst constructs. In contrast with previous studies, a newly proposed methodology that involves a dual-stage analysis of SEM and ANN was performed to provide linear and non-compensatory relationships amongst constructs. Consequently, numerous distinct types of studies in diverse sectors have conducted hybrid SEM–ANN analysis. Accordingly, the current work supplements the academic literature with a systematic review that includes all major SEM–ANN techniques used in 11 industries published in the past 6 years. This study presents a state-of-the-art SEM–ANN classification taxonomy based on industries and compares the effort in various domains to that classification. To achieve this objective, we examined the Web of Science, ScienceDirect, Scopus and IEEE Xplore® databases to retrieve 239 articles from 2016 to 2021. The obtained articles were filtered on the basis of inclusion criteria, and 60 studies were selected and classified under 11 categories. This multi-field systematic study uncovered new research possibilities, motivations, challenges, limitations and recommendations that must be addressed for the synergistic integration of multidisciplinary studies. It contributed two points of potential future work resulting from the developed taxonomy. First, the importance of the determinants of play, musical and art therapy adoption amongst autistic children within the healthcare sector is the most important consideration for future investigations. In this context, the second potential future work can use SEM–ANN to determine the barriers to adopting sensing-enhanced therapy amongst autistic children to satisfy the recommendations provided by the healthcare sector. The analysis indicates that the manufacturing and technology sectors have conducted the most number of investigations, whereas the construction and small- and medium-sized enterprise sectors have conducted the least. This study will provide a helpful reference to academics and practitioners by providing guidance and insightful knowledge for future studies.

The COVID-19 pandemic caused by the novel coronavirus SARS-CoV-2 occurred unexpectedly in China in December 2019. Tens of millions of confirmed cases and more than hundreds of thousands of confirmed deaths are reported worldwide according to the World Health Organisation. News about the virus is spreading all over social media websites. Consequently, these social media outlets are experiencing and presenting different views, opinions and emotions during various outbreak-related incidents. For computer scientists and researchers, big data are valuable assets for understanding people’s sentiments regarding current events, especially those related to the pandemic. Therefore, analysing these sentiments will yield remarkable findings. To the best of our knowledge, previous related studies have focused on one kind of infectious disease. No previous study has examined multiple diseases via sentiment analysis. Accordingly, this research aimed to review and analyse articles about the occurrence of different types of infectious diseases, such as epidemics, pandemics, viruses or outbreaks, during the last 10 years, understand the application of sentiment analysis and obtain the most important literature findings. Articles on related topics were systematically searched in five major databases, namely, ScienceDirect, PubMed, Web of Science, IEEE Xplore and Scopus, from 1 January 2010 to 30 June 2020. These indices were considered sufficiently extensive and reliable to cover our scope of the literature. Articles were selected based on our inclusion and exclusion criteria for the systematic review, with a total of n = 28 articles selected. All these articles were formed into a coherent taxonomy to describe the corresponding current standpoints in the literature in accordance with four main categories: lexicon-based models, machine learning-based models, hybrid-based models and individuals. The obtained articles were categorised into motivations related to disease mitigation, data analysis and challenges faced by researchers with respect to data, social media platforms and community. Other aspects, such as the protocol being followed by the systematic review and demographic statistics of the literature distribution, were included in the review. Interesting patterns were observed in the literature, and the identified articles were grouped accordingly. This study emphasised the current standpoint and opportunities for research in this area and promoted additional efforts towards the understanding of this research field.

In this study, pre-service teaching refers to teaching English as a second language (TESL) to Malaysian students whose first language is not English. TESL prepares English-language learners to become future teachers of English as a second language. To date, no multi-criteria framework has been developed to evaluate and select the skills of pre-service teachers. This study presents a new framework to assess and rank the English skills of pre-service teachers on the basis of fuzzy Delphi and multi-criteria analysis. Three experiments were conducted. Firstly, criteria were identified from the literature review and the opinions of representative experts via the Delphi method. Secondly, 31 pre-service teachers were evaluated to determine the skills of pre-service teachers on the basis of Delphi criteria outcomes. English proficiency was tested through the English Language Testing Service and four language skill examinations. Each examination was evaluated by experts with vast experience in English teaching. Thirdly, pre-service teachers were ranked on the basis of a set of evaluated Delphi criteria outcomes through the technique for the order of preference by similarity to ideal solution (TOPSIS) method. Thereafter, the mean and standard deviation were utilized to ensure the identical systematic ranking of pre-service teachers. Findings are as follows. Twenty-five criteria from previous studies are representative as evaluated by the opinions of experts, which were gathered through interviews and a structured questionnaire. The validity of content was verified using a five-point Likert scale. With Delphi method outcomes, 14 criteria were selected and included in the final framework. The results of the proposed evaluation framework were tested on Malaysian pre-service teachers. TOPSIS is effective for solving the selection problems of pre-service teachers. In the final experiment, significant differences were recognized between the scores of groups, indicating identical ranking results.

This study proposes an evaluation and benchmarking decision matrix (DM) on the basis of multi-criteria decision making (MCDM) for young learners' English mobile applications (E-apps) in terms of listening, speaking, reading and writing (LSRW) skills. Benchmarking E-apps for young learners is challenging due to (a) multiple criteria, (b) criteria importance and (c) data variation. The DM was constructed on the basis of the intersection amongst evaluation criteria in terms of LSRW and E-apps for young learners. The criteria were adopted from a preschool education curriculum standard. The DM data included six E-apps as alternatives and 17 skills as criteria. Thereafter, the six E-apps were evaluated by distributing a checklist form amongst six English learning experts. These apps were subsequently benchmarked by utilising MCDM methods, namely, best-worst method (BWM) and technique for order of preference by similarity to ideal solution (TOPSIS). BWM was used for criterion weighting, whereas TOPSIS was employed to benchmark and rank the apps. TOPSIS was utilised in two contexts, namely, individual and group. In the group context, internal and external aggregations are applied. Mean was computed to ensure that the E-apps undergo a systematic ranking for objective validation. This study provides scenarios and a benchmarking checklist to evaluate and compare the proposed work with six relative studies. Results indicated that (1) BWM is suitable for criteria weighting. (2) TOPSIS is suitable for benchmarking and ranking E-apps. Moreover, the internal and external TOPSIS group decision making exhibited similar findings, with the best app being `Montessori' and the worst app being `FunWithFlupe.' (3) For objective validation, remarkable differences were observed amongst the group scores, which indicate that the internal and external ranking results are identical. (4) In the evaluation, the proposed DM revealed advantages over the six relative studies by 40.00%, 53.33%, 40.00%, 46.67%, 46.67% and 46.67%.

To survey researchers’ efforts in depth in learning new problems, issues, motives on research related to children’s profiles; sculpting the literature into a clear and structed taxonomy; and determining the basic attributes of this research in terms of motivation and challenges, as well as recommendations and future studies. A focused search for each article was conducted on child profiles in four major databases: ScienceDirect, Web of Science, EBSCO, and ERIC. These databases are broad and sufficient to cover child profile studies in the literature. The initial query search resulted in 99 articles: (15/99) from ScienceDirect database, (45/99) from WoS, (23/99) from ERIC, and (16/99) articles from EBSCO, from 2011 to 2016. Those papers were thoroughly perused for the main purpose of developing a general map for research conducted on this emerging topic. Most of the articles (81.82%; 81/99) were measurement and evaluation papers; (15.15%; 15/99) were review and survey papers that refer to the literature in order to describe the child profile; and (3.03%; 3/99) were a design. Since 2011, researchers have followed the trend of child profile applications in many ways, while leaving certain aspects for further attention. Regardless of their categorization, articles focus on several challenges that hinder the full utility of child profile apps and do recommend mitigations. Research on child profiles is active and highly varied. In this paper, we hope that this review of previous studies will contribute to understanding the challenges and gaps that are available to other researchers to join this research line.