Tax evasion based on related party transactions is a new strategy in tax evasion that is carried out through legal transactions, such as transactions between a group of companies that have heterogeneous, complex, and hidden interaction relationships for tax evasion. Existing studies cannot effectively identify tax evasion behaviors of related parties because the machine learning-based audit method can detect the abnormal financial status of individuals with high accuracy and efficiency. However, it is helpless when faced with heterogeneous, complex, and hidden interaction relationships and cannot identify tax evasion groups with related party transactions. The hybrid of graph mining and deep neural network approaches has the ability to detect anomalies in complex organizational structures. In this study, 1,780 companies with related party transactions, including 523 companies located in free trade zones and 1,257 companies located outside free trade zones, which have a common board member and economic activity of production or trade, were selected. In this study, financial and tax data from tax returns and the systems of the Iranian Tax Administration from 2016 to 2019 were used. This study is practical in terms of purpose. Python software and the NetworkX package were used to estimate the model. To predict tax evasion in related party transactions, three algorithms were used: Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Multilayer Perceptron Neural Network (MLP) in deep mode. To identify suspicious groups, three steps were taken; first: detecting tax rate differences, matching the topological pattern, and identifying tax burden anomalies; second: experimental tests based on data from 16,756 related party transaction purchases and sales in the country; third: estimating the coefficients and the relationship between the topological pattern in the two cases of profit retention and profit transfer based on the graph mining approach and deep neural network. The results show that both profit retention and profit shifting exist in tax evasion of related party transactions. However, based on the results, the intensity of the profit retention relationship in tax evasion of related party transactions is stronger than the profit shifting relationship. Based on the results, the graph mining approach was more accurate than the logit, probit, and linear probability models.

The insurance industry plays a vital role in economic stability and risk management. This study aims to forecast the return on assets (ROA) in Iran’s insurance sector using the Bayesian Structural Time Series (BSTS) model, while identifying key macroeconomic drivers via the TVP-Graph approach. Based on time-series data from 2011 to 2024, findings indicate that capital adequacy and financial leverage are the most influential factors on ROA, whereas unemployment has no significant effect. The 24-month forecast suggests relatively stable yet fragile financial performance, highlighting a high degree of uncertainty and the need for structural and policy reforms.

This study investigates the impact of economic sanctions on tax evasion facilitated through Related-Party Transactions (RPTs) in Iran. Utilizing a novel hybrid framework that integrates graph mining, Principal Component Analysis (PCA), and advanced fuzzy metaheuristic optimization, we analyze financial data from 1, 780 companies (2016-2020). Graph mining is employed to map and detect suspicious transaction networks, particularly those involving Free Trade Zones (FTZs). A sanctions intensity index is constructed using PCA from 10 macroeconomic variables. The core predictive modeling leverages a Jaguar-optimized Type-3 Sheffer-like Type-4 fuzzy logic system to handle data uncertainty and non-linear relationships. Results indicate that sanctions exacerbate RPT-based tax evasion, increasing its magnitude from 0. 389% to 0. 414%. The proposed Jaguar model demonstrated superior performance with 98. 8% accuracy (MSFE: 0. 012), significantly outperforming traditional detection methods. Post-sanctions network topology analysis revealed a marked increase in suspicious clusters and nodes, with prevalent evasion patterns including multi-layer transfer pricing and abnormal profitability in FTZ subsidiaries. This research offers a robust, scalable tool for tax authorities to prioritize audits and enhances the understanding of how macroeconomic shocks influence illicit financial behaviors within corporate networks. IntroductionEconomic sanctions are coercive measures imposed by states to restrict international activities of target nations, offering a lower-risk alternative to military conflict (Cordesman et al., 2011). Iran exemplifies this, facing escalating sanctions that incentivize tax evasion through Related-Party Transactions (RPTs). Under sanctions, firms exploit legal gaps and accrual accounting to manipulate profits (Abeysekera, 2003, Arabi et al., 2018), transforming Iran’s financial market into a complex network (Soleimani et al., 2014). Traditional analytical methods fail against such complexity, while metaheuristic models excel. Graph mining uniquely uncovers hidden dimensions in sanctioned markets by analyzing network structures and variable relationships (Hu et al., 2022), especially where information asymmetry impedes tax authorities (Iacovacci & Lacasa, 2019, Yang & Xu, 2024). RPTs occur in nested networks with non-linear relationships (e. g., shared boards, cross-ownership) (Ruan et al., 2019). Sanctions amplify complexity through layered tactics like free trade zones (FTZs) and multi-layer transfer pricing (e. g., sequential sales at non-arm’s length prices) (Chan et al., 2016, Tian et al., 2016). Non-disclosure of ~68% key RPT information (e. g., pricing logic) exacerbates tax avoidance (Barokah, 2013), enabling profit shifting to foreign affiliates and eroding tax bases (Yang & Xu, 2024). Although RPTs can be economically justified (Gordon et al., 2004a), they risk abuse for private gain (Djankov et al., 2008, Barokah, 2013). In Iran, firms use subsidiaries in FTZs (e. g., Kish, Chabahar) and transfer pricing under Article 132-T of Iran’s Direct Taxation Law to shift profits: e. g., selling goods below market to affiliates, which then export at global prices, registering profits offshore. Weak oversight and fragmented databases hinder monitoring, but Iran’s Taxpayers’ Integrated System (TIS) provides foundational data for analysis. This study proposes a novel framework combining graph mining (to detect high-risk FTZ firms) and Type-3 Sheffer-like Type-4 fuzzy logic (to model tax data uncertainty) optimized by the Jaguar metaheuristic algorithm. It identifies suspicious groups exhibiting structural (e. g., nested ownership) and behavioral (e. g., abnormal pricing) tax evasion patterns, aligning with Iran’s Comprehensive Tax Plan for risk-based audits. Research Questions: Do economic sanctions increase RPT-based tax evasion? How can advanced data analytics identify and model these hidden patterns? Theoretical Framework2. 1. Related-Party Transactions (RPTs)Per Iranian Accounting Standard 12 (Audit Organization, 2020), RPTs involve entities with control/influence over financial decisions. Key groups include: Parent/subsidiary entities under shared control. Key management personnel and relatives. Entities with significant economic/management ties. Two theoretical perspectives exist: -Agency Theory: RPTs enable opportunism by insiders (Jensen & Meckling, 1976), e. g., underpriced asset sales (Cheung et al., 2006). -Efficiency View: RPTs reduce transaction costs (Gordon et al., 2004a) but require disclosure to mitigate information asymmetry (Kohlbeck & Mayhew, 2010). Empirical evidence confirms RPTs facilitate tax avoidance via transfer pricing (Harris et al., 1993, Jian & Wong, 2010), especially in low-tax jurisdictions (Barker et al., 2016). 2. 2. Sanctions’ Economic ImpactSanctions restrict input access, raise production costs (Parsa et al., 2013), contract import-reliant sectors (Caetano et al., 2023), and reduce total factor productivity (Nosratabadi, 2023). They incentivize shifting activities to the informal economy, causing technical inefficiency (Markus, 2024). Methodology3. 1. Data & Variables-Dependent Variable: Tax evasion, measured by the tax gap (difference between declared and final tax) per OECD standards (Slemrod & Weber, 2012). -Independent Variable: RPT volume (Iranian Accounting Standard 12). -Moderator: Sanctions index (PCA-derived from 10 macroeconomic variables, Table 1). Data: 16, 756 RPTs from 1, 780 Iranian firms (2016–2020), including: 523 firms in FTZs (zero tax rate under Article 132-T). 1, 257 non-FTZ firms with shared boards. Financial data (net sales, COGS, operating profit) sourced confidentially from Iran’s National Tax Administration (INTA). 3. 2. Integrated FrameworkGraph Mining: Construct transaction networks (nodes = firms, edges = RPTs weighted by price deviation). Identify high-risk clusters(e. g., firms in FTZs with below-market pricing). PCA for Sanctions Index: -Combine 10 macroeconomic variables (e. g., oil exports, currency volatility) into a unified index. -2 principal components explain 85% variance (Table 1, Chart 3). Fuzzy Metaheuristic Optimization: -Apply Type-3 Sheffer-like Type-4 fuzzy logic to model data uncertainty (e. g., transfer pricing discrepancies). -Optimize via Jaguar algorithm (multi-objective: minimize prediction error [MSFE], maximize detection accuracy). -Output: Dynamic risk index (transaction volume, price deviation, geographic concentration). Results & Discussion-The analysis confirmed that sanctions significantly intensified RPT-based tax evasion, elevating its level from 0. 389% (pre-sanctions) to 0. 414% (post-sanctions). This 0. 025% increase, though seemingly small, represents a substantial rise in hidden economic activity within the constrained environment. -The Jaguar model achieved 98. 8% accuracy (error rate: 0. 012), outperforming traditional methods (40% vs. 74. 6% detection rate). -Graph analysis revealed post-sanctions topological shifts: increased suspicious nodes/clusters (Chart 4). -Key evasion patterns: -Multi-layer transfer pricing (e. g., mother → FTZ subsidiary → export). -Abnormal profitability in FTZ subsidiaries. -Geographic concentration in low-tax areas. Conclusion & Policy Implications5. 1. Key FindingsSanctions intensify RPT-based tax evasion by incentivizing complex, hidden transaction networks. The integrated graph-fuzzy-jaguar framework proves superior to linear models in detecting evasion under data uncertainty. 5. 2. Innovations-First application of Type-3 fuzzy logic in taxation. -Dynamic risk index for audit prioritization. -Operational compatibility with INTA’s existing systems (e. g., TIS). 5. 3. Recommendations-To INTA: Integrating the model into a blockchain-based real-time monitoring platform and Develop an AI dashboard with risk-tiered visualization (green/yellow/red). -Domestic Policy: Mandating disclosure of transfer pricing logic and topological RPT networks and establishing a National Networked Data Analysis Center. -International Cooperation: Leveraging double-taxation agreements for cross-border data exchange. -Future Research: Extending the model to multinational contexts and designing "tax resilience indices" for sanction-affected economies.

Nowadays, whereas the use of social networks and computer networks is increasing, the amount of associated complex data with graph structure and their applications, such as classification, clustering, link prediction, and recommender systems, has risen significantly. Because of security problems and societal concerns, anomaly detection is becoming a vital problem in most fields. Applications that use a heterogeneous graph, are confronted with many issues, such as different kinds of neighbors, different feature types, and differences in type and number of links. So, in this research, we employ the HetGNN model with some changes in loss functions and parameters for heterogeneous graph embedding to capture the whole graph features (structure and content) for anomaly detection, then pass it to a VAE to discover anomalous nodes based on reconstruction error. Our experiments on AMiner data set with many base-lines illustrate that our model outperforms state-of-the-arts methods in heterogeneous graphs while considering all types of attributes.

The independence graph Ind(G) of a graph G is the graph with vertices as maximum independent sets of G and two vertices are adjacent, if and only if the corresponding maximum independent sets are disjoint. In this work, we find the independence graph of Cartesian product of d copies of complete graphs Kq, which is known as the Hamming graph H(d, q). Greenwell and Lovasz [7] found that the independence number of direct product of d copies of Kq as qd−1. We prove that the independence number of Hamming graph H(d, q), which is cartesian product of d copies of Kq, is also qd−1. As an application of our findings, we find answers for rook problem in higher dimensional square chess board.

In this paper, we find the star chromatic number of central graph of complete bipartite graph and corona graph of complete graph with path and cycle.