The conventional e-commerce business chain is undergoing a transformation centered on short videos and live streams, giving rise to interest-based e-commerce as a burgeoning trend in the industry. Varied content stimulates the fast growth of interest in e-commerce. By employing wireless sensor networks (WSNs) to collect real-time data on user behavior, preferences, and contextual factors, businesses employ high-tech analytics and predictive modeling systems to evaluate individual purchasing power. This new integration supports E-commerce platforms to offer personalized and targeted product recommendations, pricing strategies, and promotional campaigns, thus optimizing the customer shopping experience. The WSN-assisted predictive abilities not only allow businesses to tailor their offerings to particular user segments for contributing to the overall performances and effectiveness of E-commerce ecosystems in a gradually dynamic market. This study develops a WSN-Assisted Consumer Purchasing Power Prediction via Barracuda Swarm Optimization Algorithm Driven Deep Learning (CP3-BSOADL) for E-Commerce Systems. The major aim of the CP3-BSOADL technique is to precisely forecast the procuring power level with the customer content preferences to offer new concepts for interest e-commerce systems. In the CP3-BSOADL technique, two major processes are involved. For the prediction process, the CP3-BSOADL technique utilizes a stacked auto-encoder (SAE) model which effectually forecasts the purchasing power of the consumers for e-commerce systems. Besides, the BSO algorithm can be applied to effectually fine-tune the hyperparameters related to the SAE model which leads to accomplishing enhanced predictive results. The performance analysis of the CP3-BSOADL technique is tested using an e-commerce dataset. The extensive result analysis stated that the CP3-BSOADL technique gains better performance over other recent state-of-the-art approaches in terms of distinct measures.

中文翻译：

---

通过梭子鱼群优化驱动的电子商务系统深度学习，WSN 辅助消费者购买力预测