Research Paper
Management approaches in the field of smart
Somayeh Akhavan; Mehdi Elyasi; Soroush Ghazinoori; mehdi goodarzi
Abstract
Innovation in the food industry is increasingly dependent on networked interactions among firms, universities, and other innovation actors. However, in emerging economies, the formation and sustainability of these networks face multiple institutional and cultural challenges. This study aims to develop ...
Read More
Innovation in the food industry is increasingly dependent on networked interactions among firms, universities, and other innovation actors. However, in emerging economies, the formation and sustainability of these networks face multiple institutional and cultural challenges. This study aims to develop a conceptual framework for the formation and sustainability of innovation networks around food industry firms in Iran, with a focus on the mediating role of technology. The research adopts a qualitative approach within an interpretivist paradigm, and data were collected through 31 semi-structured interviews with academic experts, industrial managers, and intermediary actors. Data analysis was conducted using multilevel thematic analysis (open, axial, and selective coding). The findings indicate that innovation networks in Iran's food industry are predominantly informal, short-term, and fragile, influenced simultaneously by weak collaboration culture, organizational logic imbalances, secrecy regimes stemming from low technological levels and ease of copying, as well as political instability and low institutional trust. Results also show that technological intermediaries, despite their potential to reduce transaction costs, build trust, and align actors, play an unstable and marginal role due to weak network governance and the absence of sustainable credibility mechanisms. Based on the extracted mechanism chains, the study proposes a phased, trust-based conceptual framework for developing innovation networks, emphasizing the implementation of low-risk pilots, professionalization of intermediaries, institutional transparency, and the design of mutually beneficial mechanisms. be in past tense.)
Research Paper
Management approaches in the field of smart
Mahmoud Zahedian Nezhad; Mohammad Mehraeen; Rouhollah Bagheri; Seyyed Mohammad Tabatabaei
Abstract
Cardiovascular Diseases (CVDs) represent a primary cause of global mortality. The proliferation of complex data from diagnostic tools like ECG poses significant challenges for clinicians, affecting diagnostic accuracy and delaying treatment. While Ensemble Learning (EL) offers enhanced performance by ...
Read More
Cardiovascular Diseases (CVDs) represent a primary cause of global mortality. The proliferation of complex data from diagnostic tools like ECG poses significant challenges for clinicians, affecting diagnostic accuracy and delaying treatment. While Ensemble Learning (EL) offers enhanced performance by integrating multiple models, a systematic comparison of its techniques within CVD management has been limited. This study utilizes a meta-synthesis to investigate the application of EL models, often combined with Machine Learning (ML) and Deep Learning (DL). The research aims to categorize EL models in CVD management, evaluate their performance, identify their advantages and limitations, and analyze the role of feature engineering. Our findings show that EL applications are classified into four domains: prediction, diagnosis, identification, and classification. The results confirm EL models are dominant across all categories, with their effectiveness heightened when integrated with ML and DL. Notably, Random Forest (RF) and gradient boosting models like XGBoost are the most frequently implemented and highest-performing techniques, consistently yielding superior results. This study offers valuable insights for researchers and clinicians, providing a framework for applying hybrid models to achieve more precise and effective management of cardiovascular diseases.
Research Paper
Data science, intelligence and future analysis
Homa Khodadadi; Mostafa Kzaemi; Naser Motahari Farimani; Seyyed Mohammad Tabatabaei
Abstract
Migration of human resources in the health sector not only reduces the quality of healthcare services but also imposes detrimental social and economic consequences on developing countries such as Iran. Therefore, accurately modeling the decision-making behavior of this workforce requires advanced analytical ...
Read More
Migration of human resources in the health sector not only reduces the quality of healthcare services but also imposes detrimental social and economic consequences on developing countries such as Iran. Therefore, accurately modeling the decision-making behavior of this workforce requires advanced analytical approaches to capture complexities and social interactions. This study aimed to design and validate a data driven agent-based model to simulate migration behavior among healthcare professionals in Iran. Secondary data were employed from the 2023 survey entitled “National Survey on Elite Migration and Factors Influencing the Outflow of Human Capital in the Health Sector” conducted by the Iranian Migration Observatory using a standardized questionnaire. The research adopted a hybrid framework in which 384 balanced samples were used for training, and the Random Forest machine learning algorithm served as the behavioral meta model of agents to directly extract nonlinear decision-making rules from microdata. The model output, representing the migration probability of each agent, was then integrated into the agent-based simulation, where comparison with an optimal decision threshold determined the final migration or non-migration action. Results indicated that the data driven agent model significantly outperformed the theory driven agent model based on logistic regression in predicting migration intentions. Furthermore, analyses confirmed that key variables such as age, work experience, and social network effects played nonlinear and essential roles in shaping final decisions.
Research Paper
Data science, intelligence and future analysis
Mohammadreza Kazemi; ,Alireza Dehghanpour-Farashah; Afsaneh Dehghanpour-Farashah
Abstract
The rapid emergence of Artificial Intelligence (AI) and novel digital technologies has initiated a fundamental transformation in the public governance sphere, yet a profound gap is observed between their technical capabilities and practical application within governmental organizations. This research ...
Read More
The rapid emergence of Artificial Intelligence (AI) and novel digital technologies has initiated a fundamental transformation in the public governance sphere, yet a profound gap is observed between their technical capabilities and practical application within governmental organizations. This research aims to conduct a futures study on the smart public governance of government human capital and delineate the strategic trajectory for intelligent transformation in light of digital evolution. The body of literature selected for the meta-synthesis included all scholarly articles and reports published between 2020 and 2025 addressing AI in the public sector, digital transformation, and data-driven governance. In the first phase, the meta-synthesis method was employed to systematically identify the key uncertainties and main drivers of governance transformation. In the second phase, the probable scenarios for the future of smart public governance were developed. The research findings highlighted the complex dimensions of this transformation, which include the pressing need for the public sector to acquire human resources with digital competencies, as well as the necessity of establishing coherent public governance frameworks to manage risks and algorithmic biases.
Research Paper
Management approaches in the field of smart
meysam davoodi; Seyed Ehsan Zahouri; bahram alishiri
Abstract
This study aimed to develop a comprehensive Artificial Intelligence (AI)-based policy implementation model for the Central Branch of the Social Security Organization (SSO) of Khuzestan Province. Given the extensive scope of the SSO’s services, leveraging AI is essential for enhancing transparency, ...
Read More
This study aimed to develop a comprehensive Artificial Intelligence (AI)-based policy implementation model for the Central Branch of the Social Security Organization (SSO) of Khuzestan Province. Given the extensive scope of the SSO’s services, leveraging AI is essential for enhancing transparency, execution efficiency, and accountability in response to complex socio-economic challenges.Methodologically, this research is applied in objective and mixed-methods (qualitative-quantitative) in design. The qualitative phase utilized the Meta-Synthesis method to extract the model’s components. The sample comprised 30 scientific articles and documents (domestic and foreign), which, upon coding, led to the identification of 10 main components and 53 indicators.In the quantitative phase, the Fuzzy Delphi method was employed for validation, confirmation, and weighting of the components. The sample for this phase included 15 academic and executive experts in the fields of AI and Social Security.The Meta-Synthesis results indicated that the AI policy implementation model consists of 10 key components, the most significant of which include: “Analytical Capacity and Decision Support,” “Human Resources Requirements and Skills,” “Technology and Operations Integration,” and “Ethical, Legal, and Security Challenges.” Furthermore, the Fuzzy Delphi findings confirmed these components and prioritized them within the SSO context, establishing final implementation requirements. The resulting model provides a comprehensive 10-component framework that introduces AI tools for enhanced efficiency while emphasizing governance and ethics to facilitate responsible and accountable policy execution.Keywords: Meta-Synthesis, Fuzzy Delphi, Policymaking, Artificial Intelligence, Social Security Organization, Policy Implementation Model.
Research Paper
Management approaches in the field of smart
Samaneh Moradi; Mehrdad Hosseini Shakib; Ali Badizadeh
Abstract
In the era of the Fourth Industrial Revolution, digitalization and implementation of emerging technologies are considered as the main drivers of transformation in various industries. The railway transportation industry is no exception to this rule and requires precise criteria to assess its maturity ...
Read More
In the era of the Fourth Industrial Revolution, digitalization and implementation of emerging technologies are considered as the main drivers of transformation in various industries. The railway transportation industry is no exception to this rule and requires precise criteria to assess its maturity level and readiness in adopting and implementing Industry 4.0 related technologies. The main objective of this research is to identify and present a conceptual model for railway transportation industry maturity assessment based on fourth-generation industrial technologies. This study was conducted using Sandelowski and Barroso's seven-stage meta-synthesis method and systematic review of 87 scientific articles published between 2016 and 2025. The article screening process was based on precise evaluation criteria including language, temporal scope, study conditions, research population, and article types. The research results led to the development of a conceptual model comprising 5 main dimensions, 21 indicators, and 84 operational codes, which include: Industry 4.0 technologies in railway transportation, digitalization challenges and barriers, cybersecurity and digital risks, practical applications and performance improvement, and sustainability and environment. The scientific validity of the results was confirmed with a Kappa coefficient of 0.89 and content validity of 0.83. This model provides a comprehensive framework for evaluating and measuring the maturity level of organizations active in railway transportation and can be used as an effective tool for identifying strengths and weaknesses, and developing improvement strategies in the digitalization path.
Research Paper
Management approaches in the field of smart
mohammad rabiei
Abstract
Semantic similarity is used in applications such as information retrieval, text summarization and sentiment analysis. In this article, a new method based on deep learning has been presented in order to check the matching percentage of the proposed name of the company registration applicants with the ...
Read More
Semantic similarity is used in applications such as information retrieval, text summarization and sentiment analysis. In this article, a new method based on deep learning has been presented in order to check the matching percentage of the proposed name of the company registration applicants with the time of the company's activity. The key innovation lies in the use of a combined Aria BERT model for word embedding to convert registered company names into vectors. Additionally, the company's field of activity is converted into numerical vectors using the FastText model, which are then processed through deep learning algorithms, specifically bidirectional long short-term memory (Bi-LSTM) networks with an additional attention layer. The results were evaluated using cosine similarity and ROUGE criteria. Following the approval of the company name and activity field, the DBSCAN clustering method is employed to categorize the company names based on their activities. The results demonstrate that the ROUGE-1, ROUGE-2, and ROUGE-L scores for company activity vectorization are 0/7623, 0/7413, and 0/7982, respectively. The overall model accuracy and recall were 0/8512 and 0/8317, respectively. Moreover, the correlation coefficient between the cosine similarity of the proposed names and the company's activity time, as calculated by the model, was 93%, confirming the model's effectiveness.This method effectively preventing the registration of names that do not meaningfully relate to the company's operations. By clustering company names, the method facilitates the suggestion of related names based on the company's field of activity.
Research Paper
Management approaches in the field of smart
maryam nooraei abadeh; shohreh Ajoudanian; sondos bahadori
Abstract
Early-stage startups face the problem of cold start, as they have limited real-world data to train AI models. This lack of data, combined with the incompatibility of generic data with specific business needs, reduces the accuracy of predictions and recommendations. Rapid changes in data and concepts ...
Read More
Early-stage startups face the problem of cold start, as they have limited real-world data to train AI models. This lack of data, combined with the incompatibility of generic data with specific business needs, reduces the accuracy of predictions and recommendations. Rapid changes in data and concepts (such as data and concept drift), the risk of forgetting prior knowledge in transfer learning, and the heterogeneous quality of user feedback are the main challenges in this area. The proposed framework is an integrated and scalable architecture that combines transfer learning and crowd intelligence. The framework consists of four parts: collection and preprocessing of (limited), generic, and user feedback real-world data; transfer learning with a pretrained model and efficient optimization to prevent forgetting prior knowledge; model enhancement with filtered and weighted user feedback; and continuous prediction by monitoring data and concept changes with mathematical criteria. The training data is composed of a combination of real, generic, and user feedback data, and optimization is performed by minimizing error and controlling complexity. Evaluation on three real datasets. Other metrics such as prediction accuracy, positive sample detection, balance between the two, error reduction, and data stability were also improved in all three datasets, especially in investment data that is more scattered. This framework increases the efficiency of limited data and ensures the stability of the model.
