Fake reviews will badly affect the authenticity of digital platforms, and it is the responsibility of the industry to ensure the factuality of the information about products and services to increase the trust of the users. This work aims to design a reliable fake review detection system by leveraging both aspect-based sentiment analysis and various linguistic features of text reviews. This method gives a detailed understanding of reviewers’ opinions, which is crucial for identifying suspicious patterns that might indicate fake reviews. In addition to aspect-based sentiment analysis, five key linguistic features are incorporated into the proposed strategy. The combination of linguistic and aspect-based sentiment analysis features can contribute significantly to detecting fake reviews with reliable performance. The integration of multi-word expressions and phrase structures addresses the sophistication of spammers’ techniques, enhancing the robustness of our approach. A set of intensive experiments with the publically available datasets, Ott and Fake reviews, have been conducted to validate the model. The outcomes of the work demonstrate that the proposed hybrid feature model could contribute significantly to machine learning methods for enhancing the performance of fake review detection.