Predicting Small Business Longevity and Competitiveness Through Self - Random Manhattan Botox - Coupled Attention Network Driven Strategic Entrepreneurship

Abstract

The dynamics between the financial, operational, strategic, and environmental factors that influence a small business' longevity and competitiveness can be extremely complex and nonlinear in nature. Current predictive approaches generally do not provide adequate accuracy or adaptability over time by capturing these multidimensional dependent variables. Consequently, the purpose of this research study is to develop a new deep learning framework, called Self-Random Manhattan Botox Coupled Attention Network (Self-RanMB-CAN), for the purpose of generating robust predictions of small business longevity and competitiveness. This framework is used in conjunction with structured datasets called the Small Business Longevity Dataset, which contains 500 records, and captures 12 key features of each small business. In preparation for using the proposed framework with these datasets, CCFA (Cumulative Curve Fitting Approximation) is applied to preprocess the data to normalize and ensure the consistency of trends within the dataset prior to performing feature extraction using DCKTKT (Discrete Cosine-Krawtchouk–Tchebichef Transform) to maintain both the spectral and spatial characteristics of each feature. A combination of an RCNN (Random Coupled Neural Network) and a Manhattan Self-Attention (MaSA) framework provides greater attention to business variables that are most impactful and thus improve feature selection. In addition, the Botox Optimization Algorithm (BOA) optimizes network parameters, which increases convergence stability. The experimental results show a classification accuracy of 99.9%, along with high precision, recall, and F1-scores; thus, the proposed framework significantly outperformed several benchmark models. The research results indicate that the proposed system can accurately model complex business dynamics and facilitate reliable predictive decision-making. In summary, the Self-RanMB-CAN is a scalable and accurate intelligent decision-support tool for strategic entrepreneurship and helps to provide a better understanding of business survival and competitive behavior in an uncertain marketplace.