Physicochemical and rheological characteristics of alcohol-free probiotic boza produced using Lactobacillus casei Shirota: Estimation of the apparent viscosity of boza using nonlinear modeling techniques


Creative Commons License

Öztürk I., Karaman S., TÖRNÜK F., SAĞDIÇ O.

Turkish Journal of Agriculture and Forestry, cilt.37, sa.4, ss.475-487, 2013 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 37 Sayı: 4
  • Basım Tarihi: 2013
  • Doi Numarası: 10.3906/tar-1207-49
  • Dergi Adı: Turkish Journal of Agriculture and Forestry
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.475-487
  • Anahtar Kelimeler: Alcohol-free, Boza, Lb. casei Shirota, Modeling, Probiotic, Rheology
  • Sivas Cumhuriyet Üniversitesi Adresli: Hayır

Özet

The effects of fermentation with 2 different starter culture mixtures composed of lactic acid bacteria (LAB) on some physicochemical, microbiological, sensorial, and rheological characteristics of boza, a traditional fermented Turkish beverage, during storage for 10 days at 8 °C were investigated. Lb. casei Shirota adapted in boza well and provided the boza with a probiotic property. At the beginning of storage, the Lb. casei Shirota count of boza was 6.83 log cfu mL-1, while total LAB counts were 8.01 and 8.11 log cfu mL-1 for S1 and S2 samples, respectively. A decrease in pH values was observed when the counts of LAB and Lb. casei Shirota increased during the storage period. Sensory acceptability scores were high for all boza samples but extending storage caused a decrease in sensory scores. Adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN), 2 important nonlinear modeling techniques, were used to construct the predictive models for the estimation of apparent viscosity during storage. All boza samples showed pseudoplastic flow behavior. Apparent viscosity decreased with increasing of shear rate, storage period, and measurement temperature. ANFIS showed better fitting performance with a higher coefficient of determination (R2 = 0.995) compared to ANN (R2 = 0.980). © TÜBİTAK.