clustering

Food insecurity can overshadow vital values and even challenge the national security of countries. Therefore, in the effort to improve food security, it is necessary for the policy maker to manage this issue with a conscious approach and a systematic method. In this regard, knowing the current state of food consumption and its determinants is not only a necessary condition for preparing future programs, but it is necessary to continue this work with the aim of monitoring and evaluating the results of implemented programs and actions. With this approach, the present study investigated the pattern of food consumption and the factors affecting it in the urban areas of Iran in 2022. To attain these objectives, the cost-income data from the Iranian Statistics Center was utilized. Additionally, the nutritional performance matrix was employed, along with the classification system provided by the Iranian Statistics Center. To better comprehension regarding the food consumption in urban regions across the provinces of the country and to pinpoint provinces exhibiting comparable behavioral patterns, the K-means clustering technique was employed. Subsequently, a map illustrating the food consumption patterns in urban areas of Iran was designed. Finally, by using the matching method, factors affecting food consumption in urban areas of Iran were identified. According to the results, the current food pattern of households in the urban areas of Iran mainly includes all kinds of cereals and this product group provides nearly 58% of the daily energy requirement. According to the clustering results, a grand total of five distinct behavioral patterns were detected. Examining the patterns shows that the pattern of food consumption in the urban areas of different provinces of the country is heterogeneous and while having a lot of diversity, it has little relationship with the geographical location of the provinces. Finally, based on the results of the matching method the factors that explain the difference in food consumption between the province include the size of the household, the age of the head of the household, the literacy level of the head of the household, the employment status of the head of the household, the average daily income of the household, the average daily subsidy received by the household, and the food variety.

Monetary and financial institutions play an essential role in the economic development of any country. Financial systems can make an economy more productive by concentrating scarce resources and funds for massive investments. Examining the bank's financial performance and creating a suitable clustering system based on indicators that affect the financial performance of banks is of great importance for bank supervisors, bank depositors and shareholders, and banking sector policymakers. In the current research, the clustering of the member banks of Iran's capital market is done based on credit risk and indicators affecting financial performance using data from the period of 2009 to 2021 related to eleven selected banks of Iran's capital market and the support vector regression (SVR) model. Two measures of return on assets (ROA) and return on equity (ROE) have been implemented and analyzed as indicators of the bank's financial performance. For this purpose, the coefficients of the banks were first extracted using the support vector regression model and then clustered with these coefficients using the average linkage method. The results indicate that the clustering of banks using both financial performance measures is similar. Based on this, in the clustering with three clusters, Saderat, Mellat, Parsian, Postbank, Pasargad, Sina, Saman, Ekhztannovin and Karabhan banks are placed in one cluster and trade and capital banks are placed in other clusters. In the clustering with four clusters, Saderat, Mellat, Parsian, Pasargad, Sina, Saman, Ekhztannovin and Karabhan banks are in one cluster and Postbank, Trade and Capital banks are in other clusters. In the clustering with five clusters according to the data of the examined banks, Saderat, Mellat, Parsian, Sina, Saman and Ekhtaznovin banks are in one cluster (category), Pasargad and Karabehan banks are in the second cluster, and Postbank, Capital and Tejarat banks are in other clusters.

Social marketing is one of the most widely used branches of marketing knowledge that is used in various social issues. Social marketing seeks to create behavioral changes that benefit individuals and society (Romero-Domínguez et al., 2021, p. 3). Non-profit organizations (NPOs) have faced intense competition for charitable donations due to declining government support and rapid growth (Hsu et al., 2021, p. 52728). A review of previous literature on donor behavioral loyalty indicates that donors may be loyal to more than one nonprofit organization in different ways (O'Reilly et al., 2012, p. 69). Achieving charity marketing goals like business marketing, in addition to choosing an efficient strategy, also requires determining effective techniques and tactics such as market segmentation, which can increase the loyalty of philanthropists even more. But statistics show that organizations lose up to 60% of their first responders. Therefore, an adequate understanding of the determinants of donor loyalty is necessary and essential (Zogaj et al., 2021, p. 379). The closest research to the current topic is the research of Masoodipoor (2018), which used other criteria for classification, so the topic is completely new and original. Therefore, the main question of this research is how people's loyalty is in the three indices of loyalty to the charity brand, loyalty to the distribution channel, and loyalty to the donation subject. The main purpose of this research is to measure the behavior of benefactors in charity affairs based on the components of loyalty to charity marketing and behavioral segmentation in the statistical population of Qom province.

The current research is a descriptive-survey type of quantitative research with an exploratory approach and is included in applied research. The statistical population of this research includes the Qom metropolis. Therefore, by using the available sampling method, two urban areas, three and seven, were selected and sampling was done from the people in those areas. According to Cochran's formula including the confidence factor of 95%, the value of 0.5 for p and q, and the error factor of 0.05, the sample size is at least 384 people. The method of data collection in this research is a questionnaire that evaluates the donation behavior of individuals in terms of loyalty to charity brand, loyalty to distribution channel, and loyalty to the subject of donation in different places and times through 42 questions in two forms of specialized and demographic questions. All questions were designed according to the Five-point Likert scale according to the variable; Therefore, the evaluation of options is variable. In total, 571 questionnaires were distributed to people physically and electronically, and 405 questionnaires were obtained that could be used for statistical analysis, and then the data of the questionnaires were analyzed using SPSS software and clustering technique and two-step clustering analysis.

Surveys show that from the total of 405 people who answered the questionnaire, the highest percentage of frequency is related to men with approximately 51.4 frequency, and the lowest number of responses is related to women with a frequency of 48.6. The youngest declared age is 14 years and the highest age is 75 years. The average age of the people studied in the research is about 32 years, and people with the age of 20 include the largest number of this sample. 59.5% of them were married. In terms of education, 37% of people have a bachelor's degree. The classification of people based on their job shows that the highest frequency of people in the studied sample belongs to freelancers with 24.7%. The most frequent percentage in the income classification, i.e. 25.4%, belongs to the category of people who do not have a fixed and independent income. Based on the answers provided, these people were classified into 5 separate clusters, 15.8% in the first cluster, 10.4% in the second cluster, 22% in the third cluster, 28.4% in the fourth cluster, and 23.5% in the fifth cluster. The highest level of loyalty to the charity brand includes the third cluster, the highest level of loyalty to the distribution channel includes the fifth cluster, and the highest level of loyalty to the donation topic includes the fourth cluster.

Due to the optional nature of charitable donations by people in the community, the competitive conditions between charitable organizations to earn more money reveal the necessity of attracting the attention of the audience. Attracting more attention from the target audience means identifying their behavior in the face of charitable affairs, and this information is used to predict people's future behavior and has a significant contribution to the growth of altruistic donations. The general results of this research indicate the existence of a gap between theory and practice. Based on the classic marketing process, segmentation is necessary to accurately and better identify the giving behavior of people and also to provide them with optimal marketing mixes. Segmentation causes people to be placed in categories with similar behavioral patterns so that advertising and promotion programs suitable for each segment are selected and implemented. The existence of criteria or criteria for segmenting people is one of the strategic challenges in the field of social policies. These criteria can measure the status of benefactors in different dimensions of charitable behavior. Since the loyalty component of philanthropists is very important for all charitable organizations, this research was conducted to investigate the loyalty behavior of philanthropists based on the geographical location of the Qom metropolis. Therefore, using the two-stage clustering technique, this research divided people's loyalty into three indicators: loyalty to the charity brand, loyalty to the distribution channel, and loyalty to the donation issue. The increase in their charitable contributions was expressed.

Deterring the location of the plate is the first step in the car license plate detection. The output of this step should be accurate enough and also, its calculations will be completed within a short time. The reason for this is that the output of this stage is as input in the next steps. If the location determination phase encounters an error, the function of the next steps will also be interrupted. In this paper, a new method is presented using the improved contrast of the image, the removal of non-numerical characters, analysis and clustering of plaque characters in order to determine the location of an Iranian car license plate with dark characters on a bright background. The proposed method reduces the overall complexity of the algorithm and In addition to ease of implementation; the speed and efficiency of the system determine the location of the plaque. The proposed algorithm is independent of the number of vehicle plates in the image, the size of the image, the inclination, and the full scintillation of the plaque and In contrast to brightness variations, it is largely resistant(robust). The test results on two different data sets with 67 and 492 images respectively, with an accuracy of 100 and 99.59% with an error rate of 1.5 and 1.63, and an average of 109 and 17.5 milliseconds, respectively.

In recent years, the most important philosophy agreed upon by organizations has been to create both economic, social, and environmental value in the form of the concept of sustainable management. Circular economy is a new concept to account for the sustainable management. The main purpose of this study is to analyze the performance of manufacturing industries from the perspective of the circular economy components. This research was conducted in the second half of 2020. The statistical population of this study includes the food industries in Bushehr province, which due to the limited size of the population, all of them were selected as statistical sample. The data collection tool of this research is a researcher-made questionnaire whose validity was checked by face content analysis method and its reliability was checked by Cronbach's alpha method. In this research, first, by studying the theoretical foundations and empirical background, effective components in circular economics were identified. Then, using the k-means algorithm, clustering of selected food industries was performed. The results showed that the food industries are in two industrial clusters: circular and linear. Next, the species detection function was obtained. It is suggested that linear industrial cluster managers pay more attention to energy efficiency, water consumption management, and sales of recyclable materials in order to transition to a circular economy. This research is of innovative nature in terms of developing the theoretical concept of circular economy and applying it to improve the performance of the food industry.

The knowledge-based economy is the newest pattern of production in the current era. So far, this pattern has resulted in unique achievements for a wide range of countries. This study aims to classify the provinces of Iran in terms of Knowledge-based economy. The classification of provinces based on their similarity in achieving the knowledge-based production pattern is the first step for correct and realistic planning. The same version cannot be used for different provinces. The regional knowledge-based economy index is defined in three dimensions: education, innovation, and information and communication technology, based on 15 sub-indices. The classification is based on the clustering technique, which is one of the branches of unsupervised learning. To do this, k-means and fuzzy c-means algorithms are used simultaneously to compare their results. The optimal number of clusters is calculated through the Silhouette coefficient. This coefficient also indicates the accuracy of the clustering results. Clustering based on the fuzzy c-means algorithm in 6-cluster case with a Silhouette coefficient of 0.77 is the most appropriate classification for research purposes. The results show that there is a clear discrepancy between different provinces in the context of knowledge-based economy. Tehran and Alborz are in separate clusters and are among the leading classes compared to others, while more than half of the provinces belong to backward cluster.

In the theory of human capital, which conceptually includes the two main components of knowledge acquisition and skills acquisition, the household and its characteristics -household dimensions-are considered as an important factor.According to importance of household dimensions in human capital theory,in this article,first, the statistical relationship between the size of household dimensions in the provinces of Iran and the share of literacy in 2016 is discussed by using the Kendall coefficient method. Then as the main purpose of the study, the trend of changes in literacy share index with increasing household dimension is examined and also clustering of provinces with independent component analysis (ICA) technique has been investigated. The results show that in households with no literate person and also one and two literate person are observed a negative significant relationship between the household dimension and the share of literacy in some provinces.Moreover, there are no significant relationship In households with three literate person in all provinces and positive significant in households with four or more literate person in all provinces except Yazd province.According this trend show that provinces are clustering in seven cluster in households with one literate person, in five clusters in households with two literate person, in seven clusters in households with three literate person, in seven clusters in households with four literate person and in four clusters in households without literate person.In all clusters, Sistan and Baluchestan province is in a separate cluster compared to other provinces according to changes in literacy share with increasing household size.

Systemic risk is the risk beared by an economic system because of a special organization. This means that a liquidity problem or a financial crisis in one company could trigger a chain of reactions that puts the whole market into trouble. This kind of risk was underestimated until 2008 financial crisis. Now federal regulations exist for controlling this risk of financial institutions. Among diversified methods of systemic risk evaluation, centrality measures have the most accuracy. In this paper, we apply a combined method of systemic risk evaluation using semi-central centrality and Markov clustering method. Results show outperformance of proposed method over classic criterion CoVaR.

The aim of this paper is to present a hybrid model to evaluate performance of loan portfolio of banking system regarding loan repayment status and to forecast credit status of loan applicants. At first stage, we have taken credit granting management approach in order to cluster and rank 100,224 loans granted by Karafarini Omid Fund. All the data on the loans granted to clients was extracted from core banking software of the Fund. Because of having access to this valuable and valid dataset, qualitative data collection methods are not used. In the first section of paper, a type of robust principal component analysis (ROBPCA) was utilized to classify the clients. Then, the eigenvector derived from ROBPCA was used as input to a two-step K-means clustering algorithm. Then, to propose a model to forecast credit status of applicants prior to granting loans, support vector machine (SVM) and artificial genetic neural networks were used. The results obtained from the applicants’ credit status forecasting showed that the model based on the artificial genetic neural networks with the mean-square error of 0.23 and %78 coefficient of determination leads to more accurate forecasting than support vector machine. Therefore, the proposed model for forecasting the applicants’ credit status can predict their performance with relative accurately. A new method in the form of data mining software provides credit institutions with the possibility of predicting applicants’ credit regarding loan repayments.

Tax evasion is one of the ongoing challenges for any tax system, particularly for developing countries. The purpose of collection of value added taxes, based on its inherent characteristics of self control and capacity of tracking the transactions chain, is the gradual clarification of economical transactions, and also built-up of a new source of income that is sustainable and reliable for financing the government’s budget. Furthermore, it is necessary that such clarification or transactions among the traders be dominant since the first process of purchasing the raw materials up to the production and sale of goods, so as the tax may be rightfully collected.
The business Intelligence in general, and data mining in particular, are effective tools for enhancing effectiveness and efficiency in the assessment of tax evasion. In the present research, based on the available information on VAT taxpayers returns in the years of study (2009-2014), being audited by the tax administration, and data mining methods including, classification algorithms of Decision Tree, Naive Bayes, and K-Nearest Neighbor, and clustering algorithms of K-means and K-medoids, the tax evasions were predicted; then by applying the silhouette index the validation of results were performed, and in view of analysis on the tax indices of earned clusters, they were divided into low-risk and high-risk taxpayers. The results arising from applying the classification and clustering algorithms may assist the tax administration in planning for the assessment and prevention of tax evasion.

In this paper, we combine the input-output approach with fuzzy clustering using the input-output tables of 2011 and 2006 in order to identify the sectoral weights of the Iranian economy. The outliers were separately clustered in order to eliminate their undesirable effects on clustering. The results show that key economic sectors in the manufacturing and mining (non-metallic minerals, basic metals, machinery and equipment n.e.c., distribution of gaseous fuels through mains, other mining, coke and refined petroleum products and nuclear fuels), in Iran are outliers, hence, need to be separately clustered. Moreover, the key sectors are "crude petroleum and natural gas" - "wholesale and retail, maintenance and repair of motorcycles and related parts and accessories, public administration, land transport and agriculture" for export promotion and job creation, respectively.

In this paper, we present indices by which it is possible to characterize and detect those potential users of false invoices in a given year, depending on the information of their tax payment, their historical performance and characteristics, using different types of data mining techniques. In this research first, clustering algorithms like Self- Organizing Map (SOM) and neural gas networks are used to identify groups of similar behaviors of taxpayers. Then decision trees, neural networks and Bayesian networks are used to identify those variables that are related to conduct of fraud and/or no fraud, detect patterns of associated behavior and establishing to what extent cases of fraud and/or no fraud can be detected with the available information. We utilize some information gained from tax auditors who are working in the Tax offices of Tehran and the informal unofficial statistics and anonymous questionnaire from some companies to gain primary data to detect fraud and compare different techniques of false invoices. To determine the main indexes in false invoices, we divided taxpayers to the micro and small enterprises and on the other side medium and large enterprises and examined the factors of fraud on each groups, with neural gas networks, separately. Particularly the neural gas method found that it was possible to identify some relevant variables to differentiate between good or bad behavior, not necessarily associated with the use and sale of false invoices. Kohonen’s method however, did not provide any behavioral patterns. In the case of micro and small businesses, the percentage of correctly detected fraud cases was 92%, while in the case of medium and large enterprises, this percentage was 89%.

Banks, on the one hand are involved with the challenge of inadequate cash to meet the customers’ needs and on the other hand, are reluctant to increase the costs resulting from the cash excess transfer. As a result, estimating the cash requirements of the bank's branches, according to their daily operations, which is considered as a multivariable system, is one of the most important issues in banking. In this regard, employing data mining, especially clustering methods and neural networks can help to increase the accuracy of estimating the cash required in branches. In this regard, Neural networks are considered significant in terms of flexibility, nonlinearity, greater tolerance to noise and independence from the basic assumptions about the input data.
In the present paper, 20 branches of Tejarat bank have been categorized in similar clusters, during the period 21/04/2014 and 22/09/2014, according to factors such as branch grade, the type of branches in terms of deposit or facility, the number of ATMs, stand-by branches. Then, considering the clustering results and the variables related to the cash of branches such as week days, payment/ deposit subsidy/ deposit interest days, holidays and official events, as well as the amount of cash used in ATMs, the suitable structure for the neural network has been identified to estimate the required cash via the error criteria and the required cash is accordingly estimated for different clusters. The results show that the neural network, considering the clustering results, can estimate the required cash of branches in different clusters with good performance with a mean absolute error of 5%.

DEA-based clustering approach reveals the input–output relationships hidden inthe data items of input and output. DEA-based clustering approach employs thepiece-wise production functions derived from the DEA method to cluster the carcompanies. Estimate production function for each car company by input-outputdata is the benefit of this method. Thus, each car company (like Saipa) not onlyknows the cluster that it belongs to, but also checks the production function typethat it confronts. It is important for managerial decision-making in different fieldswhere decision-makers are interested in knowing the changes required incombining input resources.