Ali, ZeshanZeshanAliBrottier, Zinanone RosaireZinanone RosaireBrottierAl-Khayri, Jameel M.Jameel M.Al-KhayriDr. AZHAR MuhammadAl-Dossary, OthmanOthmanAl-DossaryAl-Dalali, SamSamAl-DalaliAlsubaie, BaderBaderAlsubaieAlmaghasla, Mustafa I.Mustafa I.Almaghasla2026-06-152026-06-152026Frontiers in Nutrition, 2026, vol. 13, article no. 1839712.2296-861Xhttp://hdl.handle.net/20.500.11861/27415Open access<jats:sec> <jats:title>Introduction</jats:title> <jats:p> This study investigates the formulation, processing stability, and storage behavior of a vinegar-based beverage derived from red date ( <jats:italic>Ziziphus jujuba</jats:italic> Mill.) vinegar produced via controlled fermentation and traditional sun-aging. To improve flavor and reduce sourness, goji berry juice and honey were added to the final formulation. </jats:p> </jats:sec> <jats:sec> <jats:title>Methods</jats:title> <jats:p>The beverage was stored at 25°C, 40°C, and 50°C for 2 months to evaluate its thermal stability. Phytochemical content and antioxidant activity (DPPH, ABTS) were measured. Sensory evaluation and volatile profiling using a 28-sensor electronic nose were performed. Random Forest modeling, hierarchical clustering, feature importance analysis, and kinetic/Arrhenius modeling were applied to assess thermal markers and phenolic degradation.</jats:p> </jats:sec> <jats:sec> <jats:title>Results</jats:title> <jats:p>Elevated temperatures, particularly 50°C led to notable deterioration in phytochemicals, including reductions in total phenolic content and antioxidant activity. Sensory evaluation confirmed optimal acceptability at 25°C, whereas higher storage temperatures caused flavor imbalance and quality deterioration. Volatile profiling revealed temperature-dependent shifts in aroma compounds. The Random Forest model achieved 95% accuracy in classifying storage conditions, and feature analysis and kinetic modeling identified key thermal markers and quantified phenolic degradation.</jats:p> </jats:sec> <jats:sec> <jats:title>Discussion</jats:title> <jats:p>Overall, the findings underscore the importance of controlled storage and demonstrate the potential of multivariate and machine learning tools for shelf-life prediction and real-time quality monitoring in thermosensitive beverages.</jats:p> </jats:sec>enGogi Berry JuiceMachine LearningQualitySensory EvaluationShelf-Life PredictionThermal StorageVinegar-Based BeveragesPeer Reviewed Journal Article10.3389/fnut.2026.1839712