Review on food quality assessment using machine learning and electronic nose system

Abstract

Quality evaluation in the food industry presents a significant challenge due to the necessity for high-cost equipment and extensive analysis to ensure that products reaching consumers are safe and of the highest quality. Existing technologies often require substantial resources, trained personnel, and complex analytical procedures, creating a demand for rapid, cost-effective solutions. Electronic nose technology is an emerging approach capable of detecting and differentiating between various aromas through an array of electronic sensors, demonstrating promising results when applied to diverse food items. Machine learning algorithms play a crucial role in analyzing the complex data collected by electronic nose systems, enabling accurate identification and assessment of food based on different odors. This review explores the combination of e-nose systems with machine learning algorithms, proposing a powerful nondestructive tool for food quality assessment. By integrating advanced data processing techniques with e-nose technology, this novel approach has shown the potential in overcoming traditional limitations related to subjectivity and time-consuming analysis procedures. Furthermore, the integration of electronic noses with machine learning applications is examined across key food categories such as meat, dairy, edible oil, fish, tea, and coffee products. Various case studies are presented to highlight the efficacy of this innovative method in addressing specific quality concerns within these sectors.

Keywords

• Adulteration
• Food quality assessment
• Machine learning
• Sensors