Neutrosophic Fusion Based Rough Set Theory for Intelligent Decision Support System on Business-to-Business (B2B) Sales Estimation

Elvir Akhmetshin^1,2*, Ilyos Abdullayev³, Irina Gladysheva⁴, Emil Hajiyev⁵, Elena Klochko⁶

¹Department of Economics, Mamun University, Khiva, 220900, Uzbekistan

²Khorezm University of Economics, Urgench, 220100, Uzbekistan

³Department of Business and Management, Urgench State University, Urgench, 220100, Uzbekistan

⁴Department of Management, RUDN University, Moscow, 117198, Russia

⁵Department of Business Management, Azerbaijan State University of Economics (UNEC), Baku, AZ1001, Republic of Azerbaijan

⁶Department of Management, Kuban State Agrarian University named after I.T. Trubilin, Krasnodar, 350044, Russia

Emails: akhmetshin@mamunedu.uz; ilyos.a@urdu.uz; gladysheva-iv@rudn.ru; hajiyev.emil@unec.edu.az; klochko.e@edu.kubsau.ru

Abstract

One of the most effective devices to model uncertainty in decision-making difficulties is the Neutrosophic set (NS) and its extensions, like interval NS (INS), interval complex NS (ICNS), and complex NS (CNS). Predicting the result of sales benefits is the essential element of effective business management. Traditionally, undertaking this prediction has depended generally on individual human analyses in the sales decision-making process. A model of business-to-business (B2B) sales predicting is a difficult decision-making procedure. There are several methods for supporting this procedure; however, generally it is even established on the individual judgments of the decision-maker. The B2B sales predicting problem is represented as the prediction problem. Presently, intelligible predictive methods were analyzed and studied utilizing the technique of machine learning (ML) to increase the upcoming sales prediction. This paper presents an Adaptive Intelligent Business to Business Sales Estimation using Neutrosophic Fusion of Rough Set Theory (AIB2BSE-NFRST) model. The main intention of AIB2BSE-NFRST technique is to enhance prediction analysis for B2B sales estimation using advanced techniques. Initially, the data pre-processing performs min-max normalization to prepare raw input data for analysis by transforming it into a structured format. Furthermore, the proposed AIB2BSE-NFRST technique utilizes NFRST method for the prediction process. To further optimize model performance, the seagull optimization algorithm (SOA) is utilized for hyperparameter tuning to ensure that the best hyperparameter is selected. To exhibit the enhanced performance of the presented AIB2BSE-NFRST model, a comprehensive experimental analysis is made under the E-commerce sales dataset. The AIB2BSE-NFRST model outperforms existing techniques with a superior MSE of 0.0033, highlighting its exceptional accuracy in B2B sales estimation.

Keywords: Business-to-Business; Sales Estimation; Neutrosophic Logic; Neutrosophic Fusion of Rough Set Theory; Neutrosophic Set; Seagull Optimization Algorithm