Data mining, fuzzy AHP and TOPSIS for optimizing taxpayer supervision

Jupri, M.; Sarno, Riyanarto

doi:10.11591/ijeecs.v18.i1.pp75-87

Cited by 12 publications

(7 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Performance measured in terms of recall and precision. Precision is the total number of positive class that is classified appropriately divided by the total data categorized as positive [25]. So, we conclude that (1356) correctly classified instances which is (90.4%) precision rate for weighted average for all possible risk levels, and (144) incorrect classified instances (i.e.…”

Section: Optimization Of Risk Assessmentmentioning

confidence: 84%

Risk assessment optimization for decision support using intelligent model based on fuzzy inference renewable rules

Radhi

2020

IJEECS

View full text Add to dashboard Cite

Due to the unreliability of wired communications and the risks of controlling the process of transmitting data besides the complications that affecting data protection and the high costs of systems infrastructure, led to use wireless communications instead of wires media, but these networks are vulnerable towards illegal attacks. The side effects of these attacks are modifying data or penetrate the security system and discover its weaknesses, which leads to great material losses. These risks and difficulties led to the reluctance of wires communications and propose intelligent techniques and robust encryption algorithms for preventing data transmitted over wireless networks to keep it safe from cyber security attacks. So, there is a persistent need for providing intelligent techniques and robust algorithms to preserve conveyed information using wireless network. This paper introduces scenario for proposing intelligent technique to increase data reliability and provides a new way to improve high level of protection besides reduces infrastructure cost. The proposed system relies on two models, where the first model based on producing a knowledge base of risk rules while the aim of the second module is a risk assessment outcomes and encryption process according to attacks type. In this system, reducing risks levels based on renewable rules whereas a novel security system established on non-periodic keys with unsystematic operations using fuzzy system. We concluded that the proposed system has the ability to protect the transmitted data, increases its reliability and reduce the potential risks. MATLAB Toolkit 2014 then Weka open source package was used in encryption and data mining for the proposed system.<iframe class="ginger-extension-definitionpopup" style="left: -91.5px; top: -78px; z-index: 100001; display: none;" src="chrome-extension://kdfieneakcjfaiglcfcgkidlkmlijjnh/content/popups/definitionPopup/index.html?title=Abstract&description=a%20concept%20or%20idea%20not%20associated%20with%20any%20specific%20instance"></iframe>

show abstract

Section: Optimization Of Risk Assessmentmentioning

confidence: 84%

Risk assessment optimization for decision support using intelligent model based on fuzzy inference renewable rules

Radhi

2020

IJEECS

View full text Add to dashboard Cite

show abstract

“…Realistic data typically takes from heterogeneous platforms and may be redundant, incomplete, and inconsistent [20]. Thus, it requires a preprocessing step that converts data into a suitable format for analysis and discovery [21].…”

Section: Data Preprocessingmentioning

confidence: 99%

BotDetectorFW: an optimized botnet detection framework based on five features-distance measures supported by comparisons of four machine learning classifiers using CICIDS2017 dataset

Jabbar

Mohammed

2021

IJEECS

View full text Add to dashboard Cite

<p><span>A Botnet is one of many attacks that can execute malicious tasks and develop continuously. Therefore, current research introduces a comparison framework, called BotDetectorFW, with classification and complexity improvements for the detection of Botnet attack using CICIDS2017 dataset. It is a free online dataset consist of several attacks with high-dimensions features. The process of feature selection is a significant step to obtain the least features by eliminating irrelated features and consequently reduces the detection time. This process implemented inside BotDetectorFW using two steps; data clustering and five distance measure formulas (cosine, dice, driver & kroeber, overlap, and pearson correlation) using C#, followed by selecting the best N features used as input into four classifier algorithms evaluated using machine learning (WEKA); multilayerperceptron, JRip, IBK, and random forest. In BotDetectorFW, the thoughtful and diligent cleaning of the dataset within the preprocessing stage beside the normalization, binary clustering of its features, followed by the adapting of feature selection based on suitable feature distance techniques, and finalized by testing of selected classification algorithms. All together contributed in satisfying the high-performance metrics using fewer features number (8 features as a minimum) compared to and outperforms other methods found in the literature that adopted (10 features or higher) using the same dataset. Furthermore, the results and performance evaluation of BotDetectorFM shows a competitive impact in terms of classification accuracy (ACC), precision (Pr), recall (Rc), and f-measure (F1) metrics.</span></p>

show abstract

“…It is specifically designed to handle situations where there are multiple criteria involved in the process of ranking alternatives and determining the most favorable option. This is achieved by employing distance measures to assess the relative performance of each alternative (Almoghathawi et al, 2017;Balcı, 2017;Behzadian et al, 2012;Chen et al, 2019;Fahami et al, 2015;Feng & Wang, 2000;Ferreira et al, 2016;Hamdan et al, 2019;Jupri & Sarno, 2019;Raed, 2020;Wasara & Ganda, 2019). The TOPSIS model is employed in this study to rank firms in the food sector and is recognized as a robust MCDM approach that yields reliable results with high computational efficiency (Hoe et al, 2020;Lukić et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Evaluating Financial Performance of Food Service Companies: A TOPSIS-based Approach

Azhar,

Abd Rahim

2024

IJARAFMS

View full text Add to dashboard Cite

The food service industry has a significant impact on society and plays a crucial role in expanding economic opportunities. Performance evaluation is essential for this industry in a highly competitive environment to enhance organizational decision-making. This study aims to propose a conceptual framework using the TOPSIS model to evaluate and rank the financial performance of food companies in Malaysia. Key financial ratios such as current ratio, return on equity, profit margin, debt to equity ratio, earnings per share, dividend yield, and price earnings ratio are employed. The study holds significance as it provides a means to assess the financial performance of the food service industry in Malaysia using the proposed conceptual framework and TOPSIS model. The findings of this research highlight that NESTLE emerges as the top-performing company among the studied companies, being the closest to the ideal solution. This paper emphasizes the importance of evaluating, comparing, and ranking the financial statuses of food service companies in Malaysia utilizing the TOPSIS model. The significance of this study lies in its ability to assess the financial performance of the food service industry in Malaysia through the proposed conceptual framework using the TOPSIS model.

show abstract

Data mining, fuzzy AHP and TOPSIS for optimizing taxpayer supervision

Cited by 12 publications

References 41 publications

Risk assessment optimization for decision support using intelligent model based on fuzzy inference renewable rules

Risk assessment optimization for decision support using intelligent model based on fuzzy inference renewable rules

BotDetectorFW: an optimized botnet detection framework based on five features-distance measures supported by comparisons of four machine learning classifiers using CICIDS2017 dataset

Evaluating Financial Performance of Food Service Companies: A TOPSIS-based Approach

Contact Info

Product

Resources

About