An Adaptive Intelligent Decision Support Framework for Business-to-Business Sales Estimation Using Generalized Q-rung Neutrosophic Soft Set

Ilyos Abdullayev^1,*, Jamshid Pardaev², Mansur Eshov³, Sanat Chuponov⁴, Elena Klochko⁵

¹Department of Business and Management, Urgench State University, Urgench, 220100, Uzbekistan

²Department of Finance and Tourism, Termez University of Economics and Service, Termez, 190111, Uzbekistan

³Department of Management and Marketing, Alfraganus University, Tashkent, 100000, Uzbekistan

⁴Department of Accounting and Business Management, Mamun University, Khiva, 220900, Uzbekistan

⁵Department of Management, Kuban State Agrarian University named after I.T. Trubilin, Krasnodar, 350044, Russia

Emails: ilyos.a@urdu.uz; jamshid_pardaev@tues.uz; m.eshov@afu.uz; chuponov_sanat@mamunedu.uz;  klochko.e@edu.kubsau.ru

Abstract

The neutrosophic set (NS) is a powerful tool for representing uncertain information in decision-making, extending conventional, fuzzy sets (FS), and intuitionistic fuzzy sets (IFS) by incorporating three degrees: truth, falsity, and indeterminacy. Sales prediction analysis wishes for intellectual data mining systems with precise predictive methods and higher trustworthiness. In the majority of cases, business depends heavily on information in addition to demand prediction of sales performance. The B2B data can offer information on how a business has to manage its products, sales team, and budget flows. Clear prediction techniques were analysed and examined using the model of machine learning (ML) to improve future sales predictions. It is challenging to manage sales prediction precision and big data (BD) when the technique of classic prediction is applied. Thus, the ML method can also be used to analyze the B2B sales reliability. This study proposes an Intelligent Business to Business Sales Estimation Framework Using Neutrosophic Soft Set (IB2BSEF-NSSS) method. The primary purpose of IB2BSEF- NSSS method is to develop an effective system for B2B sales estimation using advanced techniques for greater predictive precision. Initially, the min-max method is adopted in the data pre-processing phase to normalize input data. Additionally, the IB2BSEF-NSSS model leverages the zebra optimization algorithm (ZOA) technique for feature selection. Additionally, the generalized q-rung neutrosophic soft set (GqRNSSS) methodology is exploited for the sales prediction operation. To further increase prediction performance, the Kepler Optimizer Algorithm (KOA) model is employed for model fine-tuning, assuring optimum hyperparameter selection for upgraded accuracy. To expose the better performance of the IB2BSEF- NSSS technique, a wide-ranging experimental analysis is conducted under the B2B sales and customer insight analysis dataset. The comparison study of the IB2BSEF- NSSS technique exposed greater predictive performance, accomplishing the lowest MSE of 0.00670, indicating its efficacy over each other evaluated techniques.

Keywords: Business to Business; Sales Estimation; Generalized Q-rung Neutrosophic Soft Set; Kepler Optimization Algorithm; Fuzzy Set; Neutrosophic Set