Advanced E-Commerce Authenticity: A Novel Fusion Approach based on Deep Learning and Aspect Features for False Reviews Detection

The credibility of e-commerce platforms is increasingly threatened by the proliferation of false and deceptive reviews generated by bots, paid spammers, and malicious competitors. These reviews are often crafted to closely resemble genuine user opinions, rendering traditional machine-learning and rule-based detection systems ineffective. Although deep learning models have demonstrated improved performance by capturing semantic and contextual patterns in review text, they frequently lack interpretability and fail to identify aspect-level inconsistencies that are critical for understanding deceptive Bhagirathi’s paper proposes a novel fusion-based false review detection framework that integrates deep learning representations with aspect-based feature analysis to enhance both detection accuracy and transparency. Advanced neural models—including BERT, BiLSTM, and CNN—are employed to extract high-level semantic features from review text, while aspect-specific indicators such as sentiment–aspect consistency, product attribute relevance, linguistic cues, and reviewer behavioral patterns are simultaneously engineered. These heterogeneous features are combined within a unified predictive framework to exploit their complementary strengths. Experimental evaluation on publicly available e-commerce review datasets demonstrates that the proposed fusion approach significantly outperforms traditional machine-learning models and standalone deep-learning architectures. The results show notable improvements in accuracy, precision, recall, and F1 Score, along with stronger class separability, as indicated by ROC-AUC analysis. Moreover, the incorporation of aspect-level features enhances model interpretability by revealing specific product attributes and linguistic inconsistencies associated with deceptive reviews. The proposed framework offers a scalable, robust, and interpretable solution to strengthen review authenticity, protect consumer trust, and maintain the integrity of modern digital marketplaces.