Analyzing consumer behavior and shopping preferences using bootstrap aggregated neural regressor

Now, the comprehension of consumer behavior and buying preferences is vital for the development of the global marketplace. The complex and non-linear correlations in the consumer data were unable to be captured by the conventional Logistic Regression (LR) models. Hence, the prediction of various behaviors is difficult. Through the introduction of the Bootstrap Aggregated Neural Regressor (BANR) model, the challenge related to accurately analyzing and consumer behavior prediction was effectively addressed in this study. To overcome the limitations of existing methods, as it fails to comprehend consumer preferences, it is considered to be the main objective of the study. Several lightweight Neural Networks (NNs) trained with bootstrap sampling and Adversarial Training (AT) were employed by the BANR model. By integrating Meta-Learning and AT, consumer behavior was effectively and accurately comprehended, and a robust result was offered by this proposed method. With a high True Positive (TP), high True Negative (TN), and low False Positive (FP) and False Negative (FN), the suggested BANR model attains a remarkable accuracy of 99.28%, and the experimental outcomes revealed it. A value of 0.99 was attained by the following: precision, recall, and F1 scores. The ability of the BANR model enhanced the accuracy and reliability of consumer behavior prediction, and the valuable implications for hyperlocal marketing strategies in the business field were offered by this model.