neural network

The purpose of this study is to cluster the Iranian online shopping consumer market using artificial neural network, so that based on it, customers' needs can be better identified, the characteristics of each cluster can be determined more accurately, and the best clustering method can be chosen. And finally, the development of appropriate strategies for management, communication and better service to customers was achieved. Based on the purpose, the present research is descriptive and of the estimation and evaluation type, and in terms of practical purpose and in terms of time period, the situation of customers has been studied during the years 2021 to 2022. The statistical population included 52,403 online stores, and the present study selected 349 stores based on simple sampling. The method of analysis and classification of data is done by RFM, K-Means and self-organizing fuzzy-neural network. The technique used was the sum of squared error criteria and the Davies-Bouldin index. The findings showed: food items (they had the least delay in purchasing due to frequent needs); cosmetics (the majority of purchases were made by women); Luxury appliances (have the highest monetary value of purchase); industrial supplies and their accessories (the most purchases were made by men) and finally, sanitary supplies, detergents and clothes (have the most frequency of purchases). The results of the research have shown that the use of self-organizing neural networks along with the RFM method is the most suitable method for clustering and separating and valuing customers. Also, Kmeans+ANFIS also achieved good values, but the WRFM+ANFIS method has been more successful in this index.

Because of the non-linear fluctuations of the stock price, it is challenging to predict. Therefore, it is important to identify the characteristics that can be used to predict market behavior. Both traditional and modern methods are very important for these features. Among these features, we can mention the feelings and emotions of consumers. In this research, the effect of consumer sentiment factor in predicting market fluctuations has been investigated. For this purpose, the S&P500 index has been considered for forecasting first without considering the characteristics of consumer sentiments using BiLSTM deep neural network and then by combining the UMCSENT consumer sentiment factor. The experiments and results of this research show that the use of consumers' feelings and emotions increases the accuracy of forecasting the S&P500 index.

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).

Earthquakes are natural disasters that can cause high damage. Prediction of earthquakes is necessary to reduce damages, develop resistant infrastructure, and prevent the loss of life in human societies. This research aims to predict the magnitude of possible earthquakes in the Shahrood district of Khalkhal County using artificial neural networks (ANN). 

A multi-layer perceptron neural network was used to predict earthquake magnitude. 

The results showed that in the fault areas, especially the Kalur fault, earthquakes with a magnitude of 1-3 on the Richter scale are more likely to occur (70%). The likelihood of earthquakes with a magnitude of 4-6 on the Richter scale is moderate (26%) and the earthquakes with a magnitude of 7-10 on the Richter scale have a very low probability (4%).

In future studies, it is recommended to use machine learning models to better and more accurately predict earthquakes in the studied area.

This article examines and designs a prediction model for analyzing customer behavior using artificial intelligence algorithms and neural networks. The main goal of this research is to simulate and predict customer purchasing behaviors in dynamic and changing markets with the help of machine learning models. In this regard, first, the theoretical foundations related to consumer behavior prediction, applications of artificial intelligence in data analysis and neural network algorithms are examined. Then, different prediction models are introduced and their application in customer data analysis and business decision-making is described. Also, the challenges and opportunities of using artificial intelligence algorithms in predicting customer behavior in different industries are examined. The research findings show that the use of neural networks and machine learning algorithms can significantly improve the accurate prediction of customer behavior and the optimization of marketing and sales strategies. Finally, the results of modeling and predicting customer behavior for organizations and businesses can be used as an efficient tool in strategic decision-making and resource management.

The purpose of this research is to identify the dimensions and components of the implementation of information and communication technology policies with focus on the sixth development plan and to provide a model of the factors affecting the implementation with the help of neural network modeling and based on Giddens constructive theory. This research is a survey and based on the goal is an applied type. Data collection is based on the library and field method with the questionnaire tool. In order to extract the effective factors, an expert panel including experts in the field of communication has been formed, and the statistical population of the research in the model test section is the information and communication technology experts of Iran Telecommunication Company (810 people), of which 260 people were randomly selected as samples based on Cochran's formula. MATLAB software was used for data analysis. According to the findings, the best combination for development is when all input variables are considered at the same time, and the worst case is when the infrastructure development variable is ignored, and the most important based on network sensitivity analysis is related to infrastructure development and the least related to content provision.

IntroductionThe research’s aim is to evaluate meta-innovative methods to predict stock price behavior and introduce the most efficient method in the Iranian stock market. Due to the uncertainty in the field of investment and the multitude of variables, investors turn to the forecasting method, by which their estimates are close to reality and their errors are reduced. Forecasting the stock market has many complexities as it is affected by a number of economic and non-economic factors such as market news, political events, social chaos and so on (B. Labiad, 2016). Due to the volatility of the stock market, predicting stock price changes in daily trading is a challenging task and it is one of the most important issues in the financial world that has attracted the attention of most financial analysts and researchers. Financial markets are an attractive field for investment in order to achieve high profits but obtaining high profits in this market is not easy, because some data are dynamic, non-linear, and variable, and unstable in nature, and therefore there are many risks in this way (Lahmiri & Boukadoum, 2015). Recently, computational intelligence (CI) techniques such as artificial neural network (ANN), nature-inspired and fuzzy algorithms have been developed for stock price prediction due to their ability to handle noisy data in the financial market(Lahmiri & Boukadoum, 2015). In many cases, ANNs suffer from limitations such as local minima, slow convergence, long training time, and fitting difficulty due to the large number of parameters to adjust(Chandar, 2018). These problems have been neglected in much previous research. Therefore, the main goal of this research is to develop a forecasting model consisting of artificial neural networks and nature-inspired algorithms to investigate the above issues and improve forecasting accuracy. In this regard, this research seeks to answer this question: Can the hybrid models of neural networks and meta-heuristic algorithms provide accurate stock predictions?Methodology In this study, the stock prices of 5 listed companies in the base metals index of Tehran Stock Exchange in the period 1396 to 1398 were predicted. For this purpose, the optimal variables are selected from among 9 primary and widely used variables using feature selection methods, super-innovative algorithms of Harris and Wall's algorithms and then using Back propagation neural networks, radial base neural network and neural network with time delay Stock prices were predicted. Focusing on basic metal companies in Tehran Stock Exchange as a statistical population, 21 companies with the highest correlation coefficient were finally selected as a sample. To generalize the sample as much as possible to the society from among the companies, 5 companies that included small, medium, and large companies were selected as samples. The two symbols Zangan and Faravar were among the smallest companies and the two companies Femli and Foulad were among the largest companies, which were included in the sample as representatives of small and large companies, respectively, and the symbol of Fasmin was included in the sample as a representative of medium-sized companies. The required data: opening price, closing price, highest price, lowest price, closing price, volume of transactions, value of transactions, number of transactions and range of changes were collected from Tehran Stock Exchange Technology Management Company (www.tsetmc.com) and Codal website. MATLAB software was used to perform calculations and implement the proposed method.Results and Discussion The results show that in the symbol of Zangan, the combined models HHOBP, WOARBF and HHOTD have the highest R2 coefficient, respectively. In terms of MSE and MAE criteria, the HHOTD model has the lowest error rate and the lowest absolute average of the difference between the actual and predicted data, which shows that in this model, the predicted values are closer to the actual values. According to the Hitrate criterion, the HHOTD model has the highest correct estimation (89%) compared to the rest of the models. With these results, it can be concluded that Harris Hawk algorithm has the highest accuracy and the neural network with time delay has high efficiency.Conclusion Since linear models do not have the ability to understand and extract non-linear patterns such as stock price time series, the goal in this research was to combine artificial neural networks and meta-heuristic algorithms to accurately predict stock price trends. In this research, first, the optimal variables were selected using feature selection methods, meta-heuristic algorithms of Harris hawk and Whale. Then, by using neural networks after propagation of error, radial basis neural network and neural network with time delay, the stock price was predicted. Considering the findings, it is suggested that the Harris's hawk evolutionary method is superior to the whale's evolutionary algorithm. Proper navigation of the problem space and finding features is the reason for the superiority of Harris's hawk results compared to whale's algorithm. It was also shown in this research that networks with time delay have higher efficiency. Findings indicate that the hybrid method of Harris hawk and neural network with time delay is the most effective.  It is suggested to use a hybrid model of post-error propagation neural network, radial base neural network and time delay neural network with evolutionary algorithms such as genetics in future research. Also, it is suggested to use other predictive algorithms such as support vector regression in combination with neural network with time delay.

Energy is among the most influential components of economic, social, and political development of every country, and influences the political stability of countries alongside its significant role in supplying economic and financial resources. Compilation of energy market strategy considering critical conditions is thus of special significance. Therefore, the present study seeks to identify and prioritize energy market strategies under critical conditions (a case study of gas exportation from Iran to Turkey). For this purpose, a questionnaire was first handed to the statistical population including nine experts in the field of energy. A SWOT matrix was used in data analysis to discover the internal and external factors and the strategic position of energy in the country. BWM was then used to identify and prioritize the indices influencing the exportation of gas from Iran to Turkey considering critical conditions. The selected planning matrix was eventually formed, and the strategies were prioritized based on the total score of each strategy in the respective index. The present study compared 49 strategies derived from 10 indices, the results of which indicated the selection of 29 strategies as the priorities of gas exportation from Iran. Moreover, the five strategies of “investment in the field of energy and strategies to optimize its consumption through the private sector and formation of think tanks”, “prediction of the establishment of regional and international offices to develop gas exportation and enhance economic-political interaction with neighboring countries, especially those owning gas resources in the Caspian Sea based on energy interaction and development with these countries”, “integration of the supply chain and planning in gas extraction to exportation sectors in a complex operating under the National Gas Company to make decisions more agile", and “increasing the country’s gas production capacity and investment to develop oil and gas fields” which were the primary selected strategies can be used as a macro-level decision-making roadmap to establish Iran’s position in the energy market and provide the grounds for its improvement. Energy is among the most influential components of economic, social, and political development of every country, and influences the political stability of countries alongside its significant role in supplying economic and financial resources. Compilation of energy market strategy considering critical conditions is thus of special significance. Therefore, the present study seeks to identify and prioritize energy market strategies under critical conditions (a case study of gas exportation from Iran to Turkey). For this purpose, a questionnaire was first handed to the statistical population including nine experts in the field of energy. A SWOT matrix was used in data analysis to discover the internal and external factors and the strategic position of energy in the country. BWM was then used to identify and prioritize the indices influencing the exportation of gas from Iran to Turkey considering critical conditions. The selected planning matrix was eventually formed, and the strategies were prioritized based on the total score of each strategy in the respective index. The present study compared 49 strategies derived from 10 indices, the results of which indicated the selection of 29 strategies as the priorities of gas exportation from Iran. Moreover, the five strategies of “investment in the field of energy and strategies to optimize its consumption through the private sector and formation of think tanks”, “prediction of the establishment of regional and international offices to develop gas exportation and enhance economic-political interaction with neighboring countries, especially those owning gas resources in the Caspian Sea based on energy interaction and development with these countries”, “integration of the supply chain and planning in gas extraction to exportation sectors in a complex operating under the National Gas Company to make decisions more agile", and “increasing the country’s gas production capacity and investment to develop oil and gas fields” which were the primary selected strategies can be used as a macro-level decision-making roadmap to establish Iran’s position in the energy market and provide the grounds for its improvement.

Managing the process of attracting and transforming the status of faculty members of universities and higher education institutions is one of the important categories This article emphasizes that attracting expert and committed professors has a significant impact on university performance.. The purpose of the research is to analyze the performance of faculty members based on the inputs of the process of attracting and transforming the situation with the fuzzy data mining approach. Among the methods of performance analysis is data collection and analysis with data mining. The research method was using neural fuzzy data mining to analyze the performance of faculty members. This research deals with the design of neural fuzzy system to analyze the personality and performance of the faculty members recruited and converted. For this purpose, 16 personality data and 2 functional data of 1000 faculty members were recruited and transformed, firstly, they were collected, cleaned and reduced, and finally, the fuzzy neural network model was designed with machine learning methods, genetic algorithm and Ril algorithm. The implementation tool is MATLAB software. The research findings of designing a model with fuzzy neural network and Ril algorithm with the validity of high correlation coefficient and low MSE in evaluating the scientific and general competence of faculty members. From the results of the research, it is predicted that the faculty members have stable, influential, honest personality traits in general competence and extrovert, stable, influential, truthful in scientific competence to attract and transform the situation.

Life insurance has a very low adoption rate in Iran, mainly due to policy surrender. This research aims to analyze the individual characteristics and insurance contract features that influence the surrendering of term life insurance policies.

The study utilizes a pilot database of 35,171 policy-holders and pensioners registered by an Iranian insurance company in 2021. Data mining, deep learning, and neural network algorithms are used for analysis due to their high accuracy in prediction:

The model demonstrates desirable performance based on evaluation metrics with a 74 percent accuracy in predicting both types of surrendered and non-surrendered insurance policies. The model performs better in predicting non-surrendered insurance policies more attention is given to interpreting those results. Despite imbalanced data, the model still performs well. In the dataset, surrendered policies make up only 3 percent of the total, leading to bias towards predicting the majority class. Nonetheless, the model accurately predicts and categorizes most surrendered policies, covering 59 percent of the total 244 cases.

The results indicate that certain demographic characteristics, such as age, female gender, health surcharge, and accident risk rate, as well as specific contract characteristics, including policy term, time since start date, longer premium payment methods, higher annual increase in capital and premium, fewer covered risks, and lower benefits, are negatively correlated with policy surrender. Furthermore, the results suggest that if the insured person is the policy surrender themselves, the probability of surrender is minimized. On the other hand, if the insured person is someone else, especially distant relatives, the probability of surrender increases.

Supply chain management is a modern organizational management mode that organizes and plans information, capital flow, and business partnerships in the supply chain and requires complete business and market information (Quinn et al., 2012). However, the cost of acquiring supply chain companies and product information by traditional methods is very high. Information technology provides the power for companies to implement supply chain management and share the supply chain easily, and all companies in the supply chain can create value through information management (Shawaki et al., 2023). The utilization of intelligent approaches to predict prices and demand quantities enhances supplier delivery performance. It also refines demand forecasting accuracy, improves factory planning precision, forecasts demand for new products, and minimizes supplier risks, transportation costs, inventory, operational expenses, and time (Tirklai et al., 2021). In supply chain management, accurate forecasting of demand reflects the price. It is a critical issue that can reduce inventory costs and achieve the desired service level (Zouqaq et al., 2020). Intelligent supply chain pricing approaches can help supply chain companies to adapt the quality of their product offerings in supply chain management according to the knowledge gained (Kotsiopoulos et al., 2021). Identifying and modeling steel market fluctuations is very important in the steel industry and supply chain management. Considering the vertical chain in this industry and the interaction between the players of this industry, game theory has been used to model the optimal price. Neural network models were employed to replicate the game, as interaction and repeated gameplay are required for achieving balance among players. Taking into account Iran's unique circumstances, notably its confrontations with substantial sanctions in the metal industry, the sanctions variable was integrated as an adjusting factor in the pricing model for this sector.

This is a practical study. The research time frame for predicting steel prices and calculating the sanctions index spans from 2011 to 2020, with quarterly data. The MATLAB software was used.

Three Bayesian neural networks, support vectors, and Grassberg's anti-diffusion were used to predict the price of steel. The results showed that the Grossberg anti-diffusion model is more accurate in predicting steel prices. Next, the predicted price entered the game theory process and the Nash equilibrium point of the model was determined. According to the country's specific conditions, the sanctions variable was introduced in the game theory model. The results showed that the inclusion of sanctions in the model led to price increases and production reductions within the steel industry. The present study delved into price fluctuations resulting from shifts in supply and demand, particularly in the context of sanctions. The findings reveal that a reduction in supply coupled with escalated sanctions led to substantial price hikes, surpassing the impact of supply changes. Consequently, steel exhibits a heightened susceptibility to input constraints, where any disruption in its supply chain triggers significant price spikes, thus unsettling the market. This amplifies the sensitivity of supply chain management for steel. Consequently, a systemic and dynamic approach is essential for market regulation policies, raw material supply, transportation strategies, and warehousing considerations. It should be noted that the use of intelligent approaches and machine learning can play a significant role in coordinating such issues.

Considering that Stackelberg's approach was used in the current research, the sequence of players' entry into the game holds significance with respect to the Nash equilibrium. The development of market entry monitoring rules and regulations in this industry should be investigated because the steel industry is one of the industries that face high entry and exit costs. As a result, Policymakers and industry managers should monitor the entry and exit of players within this sector. They should endeavor to establish norms and regulations governing interactions among market participants to foster a structured and well-defined competitive environment.

This Study was performed to present a model for implementing information technology policies for The Sixth Development Plan based on the neural network approach.

Based on purpose, this research was applied and used survey method. Questionnaires were administered to collect data. In the factor identification phase, the statistical population consisted of 15 managers of Communication Company, who made the experts pannel. In the testing phase, 260 IT experts comprised the sample.

The factors affecting implementation include two main dimensions of structure and function, as well as ten subsequent dimensions (environment, policy, organization size, culture, technology, human resource development, infrastructure development, plans development, private section development, content provision) are worth of consideration. Results show that the neural network of the best structure consists of one entry layer with 5 entry variables, one hidden layer with 10 neurons, and one outgoing layer with one variable.

Originality:

 Taking conceptology into consideration, and using the mixed approach, we explained the components affecting the implementation of IT consisting of structure and function (both variables). Regarding methodology, modeling through neural networking approach can be considered as another contribution of this research. With regard to application value, this study makes it clear that the best combination for development is realized when all entry variables are considered simultaneously. Besides, the weakest scenario would be when the development of infrastructures is ignored.

One of the challenges faced by the steel industry is supply chain management. Based on this, inthe present research, based on the three scenarios of non-cooperation and simultaneousmovement (Cournot), non-cooperation and sequential movement (Stackelberg) and cooperativebehavior (collusion) will be discussed in the steel supply chain.The research method is applied in terms of purpose. The research period is seasonal data from2011 to 2020 and the software used MATLAB softwareIn this paper, a hybrid model based on artificial neural networks and game theory was presentedto help steel industry activists in determining the price level and optimal production. To predictsteel prices, three Bayesian neural networks, support vectors and cross-emission anti-emissionwere used. The results indicate that the cross-emission model of Grossberg is more accurate inpredicting steel prices. Also, the results show that by moving from the Cournot game to theStackelberg game and from the Stackelberg game to the Collusion game In the supply chain, itwill increase the price in the steel industry by 6 dollars per ton and the amount of product supplywill be in the range of 1500 to 4000 tons In other words, with the increase in the level ofcollusion in the steel market, more products have been offered in the market and at the sametime, the price level of the product will also increase, which will reduce the welfare of steelconsumers in the market.

Customer retention is always considered as the most important principle in all industries, and the insurance industry is no exception. During the recent years in the Iranian society, with the increase in the sale of life insurance policies, the retention of insurance customers has become more and more important to the managers and experts of the insurance industry so that they can keep a wide range of customers. Nowadys, creating a sense of satisfaction in life insurance customers as a management art has been noticed by insurance companies. The more customers the insurance company can keep happy, the less they worry about redemptions and exits. The main goal of this research is to implement data mining methods in predicting customer churn and identifying factors affecting customer churn in the life insurance products of one of Iran's insurance companies. The purpose of customer loss forecasting is to identify the desired class or class related to insurance policies that are suspended or canceled at the request of the policyholder before the end of the insurance coverage period.

In this paper, we have tried to classify life insurance customers based on abdication or non-withdrawal using data mining algorithms such as random forest, decision tree, logistic regression and neural network. The data used in this research include the information of life insurance policies of an insurance company in 2019 in Tehran province, which has a high and appropriate share in the portfolio of the insurance industry. To evaluate and compare these 4 methods, different criteria will be used. In the field of data mining, and in particular the problem of classification, the confusion matrix as a special tabulation makes it possible to visualize the performance of an algorithm. The confusion matrix shows how many true and false predictions have been made for each class, and based on these values, different criteria for classification evaluation and accuracy measurement can be defined.

The results of the research show that random forest, decision tree, logistic regression and neural network algorithms have high performance in predicting the class related to customer churn. Based on the results of the research, the probability of re-buying was better in women and people with high-risk jobs and older age. On the other hand, people who initially paid the insurance premium annually or chose a lower premium and a higher percentage of capital change factor and capital risk, the probability of their redemption was less.

Considering that life insurance is usually long-term and also considering the liquidity needs of customers and the current economic conditions of the society, insurance companies should pay more attention to life insurance customers. Also, they should put fidelity programs in order to keep customers on their agenda by continuously monitoring the customer's behavior during the insurance policy.

The present study was applied in terms of purpose and In terms of the descriptive research method, it is a survey with a comparative causal approach. A number of 100 companies listed on the Tehran Stock Exchange in 2014 to 2019 were selected as a sample. Data were collected from the databases of the Central Bank of Iran, Statistics Center of Iran, Codal website, and WDI website and Tehran Stock Exchange website, and then analyzed using Excel, 9Eviews, and Matlab software. In this study, the determinants of venture capital on the Tehran Stock Exchange (capital volume, export volume, company tax, disclosure index, rule of law index, industry diversity, diversity of company life cycle stages, type of property, number of divisions and subsidiaries, company capital, company, size and company age) By studying the research literature and using neural network and fuzzy logic, a predictive model was created to determine the investment risk based on the determined indicators.. The classification model was presented using neural network without SOM supervision and fuzzy logic flexibility. The evaluation parameters of design, accuracy of the proposed ensemble classifier, mean error of the squares, and coefficient of explanation. The simulation results indicated that the performance of the proposed design compared to the classification by MLP neural network method was improved and the presented model based on the training provided predicted the status of capital with good accuracy.

In recent decades, innovation management has always been considered by researchers. Their efforts to improve the performance of this process is ongoing. In this research, in order to design a model for managing innovation process routines, first, by reviewing previous studies, the relevant indicators were extracted and a model was presented after performing the fuzzy Delphi technique and the experts reached a consensus. To implement the model designed in ANFIS, a questionnaire was used (survey method). The designed rules came with an acceptable error with 40 training courses. Findings showed that the organizational factor is the most important variable and educational factor is in the last rank. Therefore, it is suggested that following measures should be taken: establishment of Iran Gas Turbine Center, stabilizing organizational routines as the main source of innovation, creating a suitable environment for cultural and social interactions, deepening innovative capabilities and increase technical and managerial skills.

Measuring the success rate of knowledge management implementation is difficult due to its long-term and indirect benefits in public organizations. Therefore, measuring knowledge management maturity is considered as a way to evaluate the success of knowledge management implementation in these organizations. In this regard, the present study aims to provide a framework for measuring knowledge management maturity in public organizations. To this end, in a four-step process, after identifying the weekly dimensions of knowledge management maturity and confirming them with the help of confirmatory factor analysis, artificial neural networks have been used to assess the importance of each dimension. Then, based on the position of each dimension in the importance-performance analysis matrix and examining the gap between the current and desired status of each of them through the pairwise average comparison test, the level of knowledge management maturity for each dimension is determined. The results revealed that the dimensions of leadership, organizational culture and knowledge management processes are in good condition while the dimensions of human resources, organizational structure and knowledge management strategy need to be improved. The ICT dimension is the only dimension in which it is necessary to adjust activities.

The growth of the Internet, social networks and e-commerce websites provide a platform for users to express their opinions. In recent years, many users have expressed their positive or negative opinions about food, service, and quality and restaurant atmosphere online. These comments are very important for the decision of other users as well as restaurants to maintain quality, product development and their brand. Sentiment analysis is a natural language processing approach and allows systematic analysis of users' opinions. Due to the importance of this issue, the purpose of this study is to present a model for analyzing the sentiment of TripAdvisor's comments about Iranian restaurants. In this research, we propose an aspect-based sentiment analysis based on a deep learning algorithm which is the standard long short-term memory neural network to extract users' sentiments about restaurants. To teach the model, 4000 comments were labeled according to four aspects in three classes of not related, positive and negative, and the study steps were done based on Crisp methodology. Accuracy for food, service, value and atmosphere were 82%, 86%, 87% and 81%, respectively. These results indicate the efficiency and acceptable performance of the model for aspect-based sentiment analysis of restaurants. Furthermore, food and atmosphere are the most important aspects for the customers of Iranian restaurants, respectively. Restaurant owners can use the developed model to gain a competitive advantage and find their strengths and weaknesses.

As the business process becomes more complex, the risk of financial statements being distorted increases with each passing day. In this regard, researchers have been looking for models to detect fraud in financial statements. The purpose of this research is to present an expanded model based on the quality characteristics of the auditor.

Benish (1997) used a combination of financial ratios and accruals to predict profit manipulation methods. Since auditors are presented as external oversight in the corporate governance structure of the company's performance, in this study the model is developed based on the qualitative char-acteristics of the auditor, which include the auditor's size, auditor tenure, reporting delay, Auditor Class and Auditor Change. The fitting of the vector machine, random forest and neural network has been used to fit the ex-tended model.

The results show that the coefficients obtained from the random forest model are 98.4% and more than the two neural network and vector model 93%. Also, the extended model is more accurate than the base model. Audit characteristics are influential in predicting fraud in financial statements and should be considered by capital market participants.

Originality / Value:

Research findings can be effective in improving the prediction of fraud in financial statements and also draw users' attention to the combination of financial statement information and the characteristics of the auditor's report in fraud prediction.