یادگیری ماشین

This research aims to investigate the effective factors in predicting lead time (LT) and create a predictive model of LT to improve sustainability and resilience for Kanban orders in the lean supply chain (LSC). The study follows the data mining (DM) method, and the dataset includes 103023 observations from the Kanban system, which were extracted in compliance with the requirements of the dataset quality indicators in the period 1402/6 to 1402/11. First, indicators affecting the LT of orders were extracted. Process mining was used to identify influential variables in high-variance processes to improve performance and accuracy. A stepwise analysis approach was used to select features for the model fitting stage. Also, tuning the parameters of non-parametric approaches was used. The predictive model uses Multiple Linear Regression, Multiple with curvature, Lasso, Elastic Net, Boosted Decision Tree, Bootstrap Random Forest, K-Nearest Neighbor, and Boosted MultiLayer Perceptron. The performance of the fitted regression models has been confirmed using R^2, RASE, and validation of the results and model. The results showed that the logistical features are effective in LT, and the Boosted Multi-Layer Perceptron is the best for predicting orders' LT with an accuracy of 96% and an error of 5.84. Using the model's predictive capability for new data in the Kanban system, the results obtained within four months have been used. The improvements from using DM capabilities in the Kanban system all express the significant impact of combining lean and machine learning (ML) tools to empower and resilient Lean Supply Chain Management (LSCM).

Predicting the performance of government Exchange-Traded Funds (ETFs) in the Tehran Stock Exchange holds a significant position due to their importance in macroeconomic decision-making and financial markets. This study aims to develop a model for predicting the returns of these funds using advanced machine learning algorithms, such as Support Vector Machines (SVM) and Linear Regression. The primary objective is to enhance economic and investment strategies at the governmental and institutional levels through accurate return analysis of these funds.

The study is applied in terms of purpose and descriptive-analytical in terms of methodology. Data for government ETFs in the Tehran Stock Exchange, covering the four years (2020–2023), were collected from the Tehran Stock Exchange Technology and Information Center (FIP Iran). Data analysis was conducted using two machine learning algorithms (SVM and Linear Regression) in R software. Due to the differences in asset composition, the ETFs "Palaayesh-e-Yekam" and "Daaraay-e-Yekam" were selected as the primary samples.

The results indicated that SVM and Linear Regression algorithms could predict the returns of government ETFs with high accuracy. Specifically, the "Palaayesh-e-Yekam" ETF, due to its less diverse asset composition and the inefficient Tehran market, exhibited lower prediction errors. The comparison of algorithm performance showed that both methods provided satisfactory accuracy and are suitable for future forecasting in this domain.

Originality/Value: 

By employing advanced machine learning algorithms, this research presents a practical model for predicting the returns of government ETFs in the Tehran Stock Exchange. The findings can be utilized in economic and investment decision-making at the governmental and institutional levels.

Earned value management is an important tool for predicting and assessing the time and progress of projects in the construction industry. The aim of this research is to propose a model that can reduce project scheduling problems by using the machine learning time series method, which has a very high accuracy. Accordingly, this study proposes new applications of recurrent neural network architecture based on long-term-short-term memory neural network forecasting models using the Python programming language. In this research, the difference between the project completion time in the real world and the value predicted by the machine will be determined. It should be noted that in the field of machine learning, the number of data is of particular importance. For this purpose, 03 ،EPC projects are reviewed along with the daily reports available in the project management database. Using the AHP software and multi-criteria decision-making method, the 03 reviewed projects are ranked based on three criteria: time, cost, and workload (project scope).

This study explores the phenomenon of deepfakes as a consequence of rapid advancements in artificial intelligence, machine learning, and deep learning technologies over the past decade. The primary objective is to analyze various methods for detecting deepfakes and examine their social and legal implications.

The research categorizes and evaluates four types of deepfake detection

deep learning-based, classical machine learning-based, statistical, and blockchain-based approaches. It also assesses the performance of these methods on different datasets.

The findings indicate that deep learning-based methods are more effective in detecting deepfakes compared to other approaches. Furthermore, the study analyzes the impact of deepfakes from multiple perspectives, including media and society, media production, representation, dissemination, audience, gender, law, and politics. The results reveal that society is currently unprepared to effectively combat deepfakes, due to a combination of technological, educational, and regulatory shortcomings.

Originality/Value: 

This research provides a comprehensive and comparative analysis of deepfake detection methods, offering valuable insights for policymakers and researchers. The study highlights the urgent need for effective strategies to address the rapidly evolving challenges posed by deepfakes in both social and legal contexts.

Most traditional fraud detection systems primarily focus on financial criteria to identify financial fraud, often overlooking the potential for fraudulent companies to engage in various types of non-financial misconduct. Recent studies have predominantly highlighted the significance of financial data as the sole indicator of fraud, neglecting the exploration of non-financial or Environmental, Social, and Governance (ESG) metrics as supplementary predictors. This research aims to enhance fraud prediction by integrating financial and ESG data through sophisticated machine learning and deep learning models. It examines the effectiveness of supervised machine learning and deep learning algorithms in detecting financial fraud over a 10-year period ending in 1401. This study innovatively demonstrates that a hybrid model, which combines financial and non-financial criteria, yields superior predictive accuracy for financial fraud than models based solely on financial data. The results of this study, addressing the first research question, indicate that among various machine learning and deep learning algorithms, the classification or bagging algorithm demonstrated superior efficiency. Furthermore, in response to the second research question, it was found that the dataset encompassing all features—integrating both financial and non-financial data—outperformed those datasets limited to either financial or non-financial data alone. The research results indicated that the bagging machine learning algorithms act the best with combined feature set including financial and ESG metrics combined. The adoption of the proposed model significantly improves the accuracy and effectiveness of fraud detection systems.

Timely delivery of medications, medical equipment, and other essential supplies is critical to patient care and can often be life-saving. Delivery delays in the healthcare supply chain can lead to increased costs and operational challenges for healthcare organizations and affect patient care and financial stability. Efficient and reliable supply chain management is critical to reduce these risks and ensure integrated performance in the health industry. This research addresses the delay in the delivery of health commodities in the global health supply chain of the United States Agency for International Development. It presents a framework based on the support vector machine technique and Bayesian optimization to predict the delivery status of health commodities. It also determines the features that have had the greatest impact in predicting the status of commodities delivery for data-driven health supply chain management.

The study's research method is design science, which presents a framework based on the support vector machine technique and Bayesian optimization to predict the delivery status of health commodities. It also compares the performance of different classification algorithms to predict the transportation status.

The results indicate that the presented framework based on the support vector machine technique and Bayesian optimization leads to a classification accuracy of 95%, outperforming other techniques to predict delivery delay. The results showed that the features of the destination country, shipping method, supplier, and production location are the most influential features in predicting the delivery status.

Using network data envelopment analysis (NDEA) models to assess the efficiency of Decision Making Units (DMUs) is a widely accepted method in academic research. An ongoing challenge in this field involves the computation and implementation of enhancement solutions within homogeneous clusters utilizing Machine Learning techniques. The primary aim of this paper is to identify the optimal clustering algorithm for a two-stage sustainable supply chain within the petrochemical industry in Iran. Subsequently, the application of NDEA within each cluster aims to ascertain efficiency levels and devise improvement strategies to facilitate a more targeted development approach for inefficient units. This paper investigates the best clustering algorithms in the area of Machine Learning by using quality measurement indicators and using Network Data Envelopment Analysis (NDEA) for measuring the efficiency of DMUs with sustainability approach. Upon examination, it has been determined that the Deep Embedded Clustering algorithm yields the most favorable results when applied to the data set. Furthermore, the comparison of the clustering result with the standard NDEA model has demonstrated the utility of clustering and comparing units in homogeneous categories for the purpose of efficiency calculation and determining the distance to the efficient frontier. This article, showed that how to find the best algorithm for two-stage supply chain clustering. Also, by comparing the effect of clustering on measuring the distance of inefficient units to the efficiency frontier, it was shown that clustering of units can play a significant role in planning to reach a practical development plan in each cluster.

The current research has been conducted with the aim of comprehensively investigating the effects of artificial intelligence on the future of educational management and drawing future prospects in the horizon of 1410. This study has collected and analyzed the opinions of 50 experts in the field of educational management, artificial intelligence and future research by using the mixed (quantitative-qualitative) method and using the fuzzy Delphi technique. The findings of the research show that artificial intelligence will have deep and extensive effects on various dimensions of educational management. The most important influential areas include personalization of education, assessment and measurement of educational progress, human resource management in the educational system, and data-based decision-making at different management levels. The main challenges identified are: ethical and privacy issues, the need for extensive retraining of manpower, and structural changes in the education system. Key opportunities also include increasing the efficiency and effectiveness of educational processes, wider access to quality education, and the possibility of lifelong learning. By providing strategic solutions to prepare the educational management system in the face of future developments, this study can help policy makers and educational managers in long-term planning.

Risk management in the supply chain is one of the important challenges in the business world that has a great impact on the performance of organizations. In this research, an intelligent model of risk management in the supply chain is designed using machine learning techniques. The purpose of this model is to identify and assess risks, predict their effects, and provide strategies for optimizing strategies based on large and complex data. By using machine learning algorithms such as decision trees, neural networks, and support vector machines, the proposed model is able to simulate and predict risks more accurately and quickly. The results of this research show that this intelligent model can be effective in improving decision-making and increasing the resilience of the supply chain. Ultimately, this model helps organizations to effectively manage risks and reduce the costs of crises. This study can be used as a basis for further research on the application of machine learning in supply chain risk management.

The aim of the current research is to provide a comprehensive model for providing a model for operational budgeting in fossil power plants based on machine learning and ground theory methods. In this research, the impact of the participatory approach in performance-based budgeting on accountability in the public sector was investigated. Budgeting on the basis of performance provides a basis for more accountability against the use of the organization's resources. In addition, issues such as environmental protection, high initial investment cost and optimal use of resources have a direct impact on the performance of the players in this market. Unfortunately, in the new competitive conditions, the traditional and classic models that assume the economic factors (producers or consumers) to be homogeneous and without interaction do not have the necessary efficiency and do not provide acceptable results. Operational investigation explains the behavior of heterogeneous production units and allows interaction and learning in a dynamic environment, but due to the breadth of the model and its complexities, an analytical solution cannot be used to obtain variables in equilibrium conditions. Due to the advancement of computing technology, the results of agent-based models can usually be evaluated with simulation methods in the context of different scenarios. Production units present their proposed production schedule for every hour to the operating entity. According to the consumption demand forecast for the next twenty-four hours and by implementing the auction mechanism, this institution determines the winning units and the market price, and then by implementing the settlement mechanism, sales and settlements are done. Considering the necessity of balance of consumption and production in the whole country as well as the geographical distribution of production and consumer units, in addition to the amount of consumption, economic restrictions and technical restrictions related to production, transmission and distribution are also considered in the auction mechanism. The findings of the research show that the existing infrastructure of performance-based budgeting does not affect financial accountability in the public sector, while it has a negative effect on operational accountability. The establishment of performance-based budgeting infrastructure and the use of a participatory approach in budgeting will increase the level of accountability of the government in the public sector.

The digital world provides many opportunities for marketers to reach customers. However, in the fast-paced world, finding new and innovative ways to advertise and sell products and services is very important. Due to the advancement of artificial intelligence and its development in the field of advertising and sales, professionals now have the tools to completely redefine the current understanding of branding, marketing, advertising and sales. The growing popularity of the Internet and the increased use of mobile devices are generating massive amounts of consumer data that feed artificial intelligence-based systems. This research is a type of mixed research with a qualitative and quantitative approach, which is a survey descriptive study in terms of its purpose, application, and in terms of data collection. The statistical population of the research was managers and experts in the field of digital marketing and IT in the field of advertising and sales, who were selected using the snowball sampling method. In the qualitative part, the tools for collecting information were library and articles review, interviews, and in the quantitative part, questionnaires. In the qualitative part of the data analysis method, using the theme analysis that was compiled with MAXQDA software and using the coding method, and in the quantitative part, the analysis method was based on Kendall's correlation test. According to the results of the research, 7 main themes, 22 sub-themes and 44 codes were discovered, which included the consequences of using artificial intelligence and machine learning in advertising and sales

 New technologies have profoundly changed the way people communicate and interact with their surroundings. These technologies affect every industry. Now, if according to the common definition, we define the supply chain as a set of interconnected activities that include coordination, planning and control of products and services between suppliers and customers. Looking at technological advances, we realize that these traditional structures are no longer self-sufficient, because digitalization has affected almost all aspects of human life, especially supply chain processes. The technologies of the fourth generation of industry represent the industrial revolution that has combined the Internet of Things with automatic systems such as artificial intelligence and its subset of machine learning, which are self-adjusting and self-learning. Such technological systems can change the supply chain from a centralized state to a distributed state, in fact, the deployment of these technologies provides distribution and decentralization for supply chains. The globalization of the economy and the increase in commercial competition have increased the importance of using innovative methods to achieve the goals of the supply chain. With the automation of processes, business activities have moved from manual operations to electronic transactions and all organizational processes have benefited from information and communication technologies. Considering that the design of most processes is at the disposal of centralized centers; There are always problems such as: poor efficiency, coordination at a low level and poor cooperation between the departments of a business unit, the emergence of distributed frameworks such as the blockchain platform and 4.0 generation technologies in addition to Organizations are helped in having complete transparency in transactions and cooperation with each other. They can share transactions on a peer-to-peer page. The purpose of research is to identify the components of the intelligent distributed supply chain and provide the structure of causal relationships for them, as well as the analysis of each of them in the framework of the presented structure. In the present research, codes and categories were first identified using the grounded method, and then the fuzzy cognitive mapping method was used to determine causal-effect relationships.In order to design the structure of the cause-effect relationships of the components of the intelligent distributed supply chain, it was extracted according to the fuzzy cognitive mapping method. Collective mapping was obtained by calculating the average of experts' opinions. According to its components, distributed decentralized supply chain can lead to improvement in information flows, promote healthcare and provide fair access to medical services. Also, assigning treatment priority to patients according to their physical condition for medical care, complying with the terms and conditions of the production line, reducing fraud, detecting authorized hazardous substances and removing drugs that have been licensed outside of the legal criteria can influence the company. High rents in the supply, prescription and treatment of the country to a great extent. The components of the correct implementation of the guidelines for hazardous drugs, the correct implementation of GMP rules, improvement in demand forecasting, correct and timely response of suppliers and suppliers, transparency and traceability, reduction in executive costs, Reducing the risk of implementing projects and carrying out contracts, behavioral data analysis algorithms, error and fraud detection algorithms, processes of identification, discovery, analysis, redesign, implementation and deployment, execution and monitoring, analysis of purchase plans and Procurement provides the possibility of tracking the information, political, monetary and back and forth flows of medicine, which represents its supply and demand among the manufacturer, government sector, patients, pharmacies, insurers, retailers of raw materials and importers. and reduces the possibility of fraud and corruption.Also, allocating treatment priority to patients according to their physical condition for medical care, complying with the terms and conditions of the production line, reducing fraud, detecting authorized hazardous substances and removing drugs that have been licensed outside of the legal criteria can influence the company. High rents in the supply, prescription and treatment of the country will be reduced to a great extent, which can be more effective for special and incurable patients, we know for certain that the required medicine for import and production, it is difficult and sometimes impossible to correctly identify and choose, with these created solutions, transparency is determined in whether the goal of the treatment program has been taken into account. Because we know that the distribution of special benefits is under the control of government officials and they define the priorities of subsidy allocation according to their decision-making power; This can be considered as one of the cases of corruption in the country's pharmaceutical industry. In the end, the improvement in the stock of raw materials and finished products, the non-issue of licenses for non-hazardous production lines in line with the production of hazardous drugs, the non-importation of domestic similar drugs, the elimination of middlemen and backdoors, the transparency of production costs. Transparency in contracts, transparency in the interests of the involved parties, transparency in payment to pharmaceutical companies, transparency in payment of pharmaceutical centers to drug dealers, transparency in pharmacy drug items, transparency in the risks after taking drugs, transparency in the way the budget is spent, The transparent and traceable payment system makes it clear whether the distribution and granting of licenses to natural or legal persons is done based on legal criteria and competently, while the government authority and decision-maker to avoid disclosure or The monopoly of the market cannot limit the transaction within the circle of friends and acquaintances, the benefit of political influence for the production of drugs and the distribution of imported drugs has been greatly reduced, the privileges have been removed from the circle of friends and acquaintances and in a competitive environment, they are given to companies. It will be real or legal that provide fair access to medical services.

The correct storage and arrangement of products in the warehouse increase efficiency in responding to requests, accelerate the identification of products, increase accessibility of items in the warehouse, make more use of available space in the warehouse, reduce the possibility of product damage, and increase flexibility. The review of studies in the field of warehousing and arrangement of products in warehouses revealed that the use of machine learning algorithms in this field is one of the important research gaps. Therefore, in this article, using machine learning algorithms, we aim to present an innovative algorithm for allocating goods to different parts of a warehouse, for which a real case study is used. The goal of categorizing products based on their characteristics is to minimize the total cost of the system. Hence, spatial clustering algorithms based on the density of applications with noise (DBSCAN), self-organizing mapping neural network (SOM), and AGNES are used. The obtained results show that SOM has better performance than DBSCAN. Also, the DBSCAN algorithm performs better than AGNES.

One of the most important security challenges in e-banking security centers is the inability of the internet to deal with attacks. These attacks are easily implemented and can be controlled locally or remotely. Most of these attacks are successful in reaching the main targets of the attack and bring the attacker to their desires. The reason for this is that there are many mechanisms for launching attacks based on the characteristics of the victim's server, which makes it impossible to provide a comprehensive defense solution against the attacks. Several strategies have been proposed to identify and deal with these attacks. In this paper, a combination of algorithm for selecting genetic features and machine learning methods, including decision tree algorithm, deep neural network and KNN, are presented. Provide guidelines for validation, the results obtained with other techniques such as machine learning techniques and combined with other optimization methods are compared and evaluated. In this research, 10% of KDD Cup 99 dataset for simulation has been used. First, in the preprocessing of data, the values of all attributes are converted to numbers, and the output characteristic values are changed to two values of zero and one. The results of the research indicate that the accuracy of the proposed strategy for detecting intruders compared to other recent methods is about 5%.

The lack of power to imagine, understand and learn ethical behavior cannot be compensated by organizational statements or traditional training. Meanwhile, digital simulation technologies such as virtual reality can help to learn moral behavior by providing a platform for imagining the results of actions in a simulated environment and strengthening people's imagination and being able to see things from the perspective of others.

The purpose of this study is to investigate the effect of using artificial intelligence in the war game on the cognitive performance of commanders and warriors in future naval battles.

The current research is exploratory in terms of qualitative methodology and meta-combination. The statistical population of the research consists of authentic qualitative studies published in the field of application of digital simulation technologies (with a focus on virtual reality and augmented reality) in social cognitive learning until 1401.

five components of unconscious indoctrination, the feeling of pervasive presence in the learning environment, increasing the cognitive processing capacity of abstract and non-experiential concepts, the ability to deeply understand the subject in the simulation environment, strengthening the motivation to learn, improving creativity in relation to the learning topic as the effect of the use of digital simulation technology Based on social cognitive learning, ethical behavior was identified in defense organizations.

digital simulation technologies will be able to create a significant change in social cognitive learning and the development of moral behavior in military organizations by providing the possibility

The purpose of the current research is to create an optimal portfolio using machine learning algorithms and fuzzy theory, which has a better return than the average return of the market (total index of the stock exchange).

In this article, the stocks of the selected companies are classified in the first stage using the two introduced algorithms. In the next step, stocks that entered the positive class are predicted for the next trading day with the help of random forest algorithm. For each company, three predictions are made, which are the inputs of fuzzy method optimization. Optimization is done with the aim of minimizing the risk with risk measures of value at risk and value at conditional risk. Shares information is five years old (daily) and its time period is from the beginning of 2017 to the end of 2021.

In the end, each of the algorithms and the risk measure used were measured and compared with the actual market return. Based on the obtained results, the CVAR risk measure has a better capability and result than the VAR risk measure, and the support vector machine algorithm has also achieved a better performance in choosing the investment portfolio.

Originality/ value: 

This research is optimized in the form of a capital sample by integrating machine learning methods and risk measures. Adding VaR and CVaR risk metrics enhances the decision-making process regarding risk reduction. Forecasting with the help of random forest and using an approach based on fuzzy theory for risk and value analysis gives the research an innovative perspective in portfolio formation. The findings provide investors and researchers with valuable insights in their search for better investment strategies.

In past years, having a consistent pricing strategy was considered logical. However, today, due to the rapid changes in the market structure, severe price fluctuations, intense competition and the changing needs of customers, having a fixed strategy may not be able to meet the needs of companies to stay in the market. The purpose of this research is to provide a solution based on machine learning so that it can provide a model based on the existing pricing strategies so that it can simultaneously benefit from the benefits of several strategies. In this research, by using the opinions of industry experts and background studies, parameters affecting the industry were extracted and data were collected according to that. The data were analyzed after refining in Python software using machine learning algorithm. The data were collected by the census method and in a period of almost five years from the Gela dairy products company. The relationship obtained from the range of data was evaluated by the Coefficient of determination and reached a number 98.6% that is desirable. This shows that the model presented in this research using the 4th degree polynomial regression algorithm can be useful for producing a competitive price that can meet the needs of customers and at the same time bring a good profit margin to the company.

The healthcare insurance industry faces a significant challenge predicting individuals' insurance costs, which are based on complex parameters such as age and physical characteristics. Insurance companies categorize policyholders into high-risk and low-risk groups to manage risks and avoid potential losses. However, the accurate estimation of costs for each individual can be a daunting task. By leveraging data science and machine learning techniques, insurance companies can improve their cost estimation accuracy and better manage risks. This approach can help insurance companies to provide more accurate insurance coverage and pricing for individuals leading to higher customer satisfaction and lower financial losses. To address this challenge, a data science and machine learning-based approach that uses ensemble learning to predict high-risk and low-risk individuals is used. The method involves several steps including data preprocessing, feature engineering, and cross-validation to evaluate the model's performance. The first step involves preprocessing the data by cleaning it, handling missing values, and encoding categorical variables. The second step generates new features using feature engineering techniques such as scaling, normalization, and dimensionality reduction. Next, ensemble learning is used to combine multiple regression methods such as logistic regression, neural networks, support vector machines, random forests, LightGBM, and XGBoost. By combining these methods, the aim is to leverage their strengths and minimize their weaknesses to achieve better prediction accuracy. Finally, the model's performance is evaluated using cross-validation techniques such as k-fold cross-validation. These techniques help to validate the model's accuracy and prevent overfitting. The proposed approach achieves an AUC of 0.73 demonstrating its effectiveness in predicting high-risk and low-risk individuals. In conclusion, the healthcare insurance industry can benefit greatly from data science and machine learning-based approaches. By accurately predicting high-risk and low-risk individuals, insurance companies can better manage risks and provide more accurate coverage and pricing for their customers. This can lead to the improvement of  customer satisfaction and the reduction of financial losses for insurance companies.