Optimized Isogeometric neural networks for Customer Behavior Path Analysis and Online Advertising Strategy

Abstract

In the dynamic landscape of online advertising, understanding and predicting customer behavior is essential for optimizing advertising strategies. Traditional methods often face challenges in effectively capturing the intricacies of customer journeys and subsequently tailoring advertising strategies. To address these challenges, this paper introduces a novel approach leveraging Optimized Isogeometric Neural Networks for Customer Behavior Path Analysis and Online Advertising Strategy (ISNN-CBPA-OAS). Initially, the data are obtained from Online Shoppers Purchasing Intention Dataset. Then, the data are provided to pre-processing phase. During the pre-processing phase, data cleaning is conducted using Learnable Edge Collaborative Filtering (LECF). Then, the pre-processing output is fed to feature extraction phase. The features are extracted using Dual-Tree Complex Discrete Wavelet Transform (DTCDWT) to extract demographic features, such as gender, age, postal code, education level, occupation. The selected features are given to Isogeometric neural networks (ISNN) for effectively identifying customer behavior such as Browsing, Shopping, Cart, and Marketing. Generally, ISNN doesn’t show some optimization adaption techniques to determine optimum parameter to offer accurate prediction. Reptile search optimization algorithm (RSOA) is proposed to enhance ISNN classifies the customer behavior accurately. The proposed technique is executed and efficacy of ISNN-CBPA-OAS technique is assessed with support of numerous performances like accuracy, recall, precision, F1-scorce and computational time is analyzed. Then, performance of ISNN-CBPA-OAS technique is analyzed with existing techniques likethe impact of digital marketing strategies on customer’s buying behavior in online shopping utilizing the rough set theory (PCC-CBPA-OAS), A Comparison along Interpretation of Machine Learning Algorithm for the Prediction of Online Purchase Conversion(CNN-CBPA-OAS),  Predicting customers' purchase behavior using deep learning(ANN-CBPA-OAS)respectively.