High performance implementation of tax fraud detection algorithm

Rad, Mehdi Samee; Shahbahrami, Asadollah

doi:10.1109/spis.2015.7422302

Cited by 7 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e dataset used in their study included 10028 records. e results showed that the fraud of taxpayers with a complementary sheet was about 57.9% [32].…”

Section: Review Of the Literaturementioning

confidence: 98%

See 1 more Smart Citation

Towards Tax Evasion Detection Using Improved Particle Swarm Optimization Algorithm

Mojahedi

Sangar

Màsdàrí

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

This paper employs machine learning algorithms to detect tax evasion and analyzes tax data. With the development of commercial businesses, traditional algorithms are not appropriate for solving the tax evasion detection problem. Hence, other algorithms with acceptable speed, precision, analysis, and data decisions must be used. In the case of assets and tax assessment, the integration of machine learning models with meta-heuristic algorithms increases accuracy due to optimal parameters. In this paper, intelligent machine learning algorithms are used to solve tax evasion detection. This research uses an improved particle swarm optimization (IPSO) algorithm to improve the multilayer perceptron neural network by finding the optimal weight and improving support vector machine (SVM) classifiers with optimal parameters. The IPSO-MLP and IPSO-SVM models using the IPSO algorithm are used as new models for tax evasion detection. Our proposed system applies the dataset collected from the general administration of tax affairs of West Azerbaijan province of Iran with 1500 samples for the tax evasion detection problem. The evaluations show that the IPSO-MLP model has a higher accuracy rate than the IPSO-SVM model and logistic regression. Moreover, the IPSO-MLP model has higher accuracy than SVM, Naive Bayes, k-nearest neighbor, C5.0 decision tree, and AdaBoost. The accuracy of IPSO-MLP and IPSO-SVM models is 93.68% and 92.24%, respectively.

show abstract

“…e dataset used in their study included 10028 records. e results showed that the fraud of taxpayers with a complementary sheet was about 57.9% [32].…”

Section: Review Of the Literaturementioning

confidence: 98%

“…Due to the recent development and large volume of data stored in tax systems, a tool is needed to process the stored data and detect fraudsters based on the information obtained from it. In this regard, some scholars used the parallel Bayesian network to detect forgers [32].…”

Section: Review Of the Literaturementioning

confidence: 99%

Towards Tax Evasion Detection Using Improved Particle Swarm Optimization Algorithm

Mojahedi

Sangar

Màsdàrí

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…A technique using statistical methods are developed in [2] for detecting VAT evasion done by Kazakhstani business firms. In [17], a parallel tax fraud detection algorithm is introduced using Bayesian networks as the means for parallelization.…”

Section: Related Workmentioning

confidence: 99%

Link prediction techniques to handle tax evasion

Mathews

Mehta

Suryamukhi

et al. 2021

Proceedings of the 3rd ACM India Joint International Conference on Data Science &Amp; Management of Data (8th ACM IKDD CODS &Am

View full text Add to dashboard Cite

Circular trading of goods is a carefully designed scam ubiquitous among fraudulent business dealers all around the world. Dealers involved in this scheme create an artificial trading network by issuing doctored sales-invoices amongst themselves without any movement of goods. In practice, it is observed that almost all cases of circular trade involve two or three dealers. Here, we work towards predicting circular trade involving three dealers. For the same, we built four different classification models consisting of feature variables tailored for predicting any plausible circular trade amongst three dealers. In particular, the logistic regression model gave the best performance among all the four different models with a prediction accuracy of 80%. Interestingly, we observe that a feature variable formed by using the personalised PageRank technique significantly improves the model over the state of the art link prediction variables. Predicting a future circular trade from a huge network of sales-transactions data is of significant importance to the tax enforcement officers. In addition to automating the process of detecting circular trading, which is manually impossible, this model helps them to target on a set of plausible evaders and take appropriate preventive measures. This model have been developed for the Commercial Taxes Department, Government of Telangana, India, using their first two quarter's tax returns dataset. CCS CONCEPTS• Applied computing → Economics; • Computing methodologies → Supervised learning by classification.

show abstract

“…The table confusion matrix shows the relationship between observed and estimated values for evaluating data classification results [24]. Accuracy is a comparison between data classified correctly with all data tested (5). Precision describes the number of positive category data that are classified correctly divided by the total data classified as positive (6).…”

Section: Classification Algorithms Performancementioning

confidence: 99%

“…The researchers [4], conducted a study investigating classic tax evasion cases using several methods aimed at classifying tax evasion behavior based on the network that has been simulated with real data. The researchers [5], conducted a study by applying parallelism techniques that aim to improve the performance of fraud detection algorithms. Another researchers [6], conducted a study to detect fraudulent tax invoices using various types of data mining techniques.…”

Section: Introductionmentioning

confidence: 99%

Data mining, fuzzy AHP and TOPSIS for optimizing taxpayer supervision

Jupri

Sarno

2020

IJEECS

View full text Add to dashboard Cite

The achievement of accepting optimal tax need effective and efficient tax supervision can be achieved by classifying taxpayer compliance to tax regulations. Considering this issue, this paper proposes the classification of taxpayer compliance using data mining algorithms; i.e. C4.5, Support Vector Machine, K-Nearest Neighbor, Naive Bayes, and Multilayer Perceptron based on the compliance of taxpayer data. The taxpayer compliance can be classified into four classes, which are (1) formal and material compliant taxpayers, (2) formal compliant taxpayers, (3) material compliant taxpayers, and (4) formal and material non-compliant taxpayers. Furthermore, the results of data mining algorithms are compared by using Fuzzy AHP and TOPSIS to determine the best performance classification based on the criteria of Accuracy, F-Score, and Time required. Selection of the taxpayer's priority for more detailed supervision at each level of taxpayer compliance is ranked using Fuzzy AHP and TOPSIS based on criteria of dataset variables. The results show that C4.5 is the best performance classification and achieves preference value of 0.998; whereas the MLP algorithm results from the lowest preference value of 0.131. Alternative taxpayer A233 is the top priority taxpayer with a preference value of 0.433; whereas alternative taxpayer A051 is the lowest priority taxpayer with a preference value of 0.036.

show abstract

High performance implementation of tax fraud detection algorithm

Cited by 7 publications

References 4 publications

Towards Tax Evasion Detection Using Improved Particle Swarm Optimization Algorithm

Towards Tax Evasion Detection Using Improved Particle Swarm Optimization Algorithm

Link prediction techniques to handle tax evasion

Data mining, fuzzy AHP and TOPSIS for optimizing taxpayer supervision

Contact Info

Product

Resources

About