View in EDS

A proposed classification method approach for binary variable data using Boolean algebra and an application to digital advertising.

Saved in:
Bibliographic Details
Title: A proposed classification method approach for binary variable data using Boolean algebra and an application to digital advertising.
Authors: EKELİK, Haydar, TEKIN, Mustafa
Source: Communications Series A1 Mathematics & Statistics; 2025, Vol. 74 Issue 2, p294-317, 24p
Subject Terms: CART algorithms, MULTIPLE criteria decision making, CLASSIFICATION algorithms, BOOLEAN algebra, TOPSIS method, DECISION trees
Abstract: In this paper, Boolean decision table (BDT) approach is proposed as a new classification technique for binary variables using Boolean algebra. Since the proposed BDT approach is similar to the decision tree methods used in classification analysis, the performance of the BDT approach is compared with the widely used decision tree methods in the literature: classification and regression tree (CART), random forest (RF), and extreme gradient boost (XGBoost) algorithms. While making the comparison, attention was paid to the classification performance of the models (classification accuracy, ROC, and PR curve) as well as the interpretability of the results obtained. The benefits and drawbacks of the proposed BDT approach were analyzed using real data from digital ads of an e-commerce company. The results of the analysis show that the BDT approach outperforms RF and CART algorithms in classification and is close to the XGBoost algorithm. The BDT approach has demonstrated greater validity in the digital advertising industry because, in comparison to the XGBoost algorithm, its results are more interpretable. Furthermore, classification performance was also compared using a future dataset from the same e-commerce company that is not included in the training or test datasets. Important target audiences were identified in addition to classification performance because target audiences are crucial to digital advertising. A multi-criteria decision-making technique called TOPSIS was used to ascertain the relative importance of the target audiences. Both the proposal of the BDT approach and the evaluation of the results of the classification algorithms using the TOPSIS method are considered to contribute to the literature in this field. [ABSTRACT FROM AUTHOR]
Copyright of Communications Series A1 Mathematics & Statistics is the property of Ankara University and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
Description
Abstract:In this paper, Boolean decision table (BDT) approach is proposed as a new classification technique for binary variables using Boolean algebra. Since the proposed BDT approach is similar to the decision tree methods used in classification analysis, the performance of the BDT approach is compared with the widely used decision tree methods in the literature: classification and regression tree (CART), random forest (RF), and extreme gradient boost (XGBoost) algorithms. While making the comparison, attention was paid to the classification performance of the models (classification accuracy, ROC, and PR curve) as well as the interpretability of the results obtained. The benefits and drawbacks of the proposed BDT approach were analyzed using real data from digital ads of an e-commerce company. The results of the analysis show that the BDT approach outperforms RF and CART algorithms in classification and is close to the XGBoost algorithm. The BDT approach has demonstrated greater validity in the digital advertising industry because, in comparison to the XGBoost algorithm, its results are more interpretable. Furthermore, classification performance was also compared using a future dataset from the same e-commerce company that is not included in the training or test datasets. Important target audiences were identified in addition to classification performance because target audiences are crucial to digital advertising. A multi-criteria decision-making technique called TOPSIS was used to ascertain the relative importance of the target audiences. Both the proposal of the BDT approach and the evaluation of the results of the classification algorithms using the TOPSIS method are considered to contribute to the literature in this field. [ABSTRACT FROM AUTHOR]
ISSN:13035991
DOI:10.31801/cfsuasmas.1502723