Journal of Agri-Food Science and Technology

Effect of Ulva sp. Incorporation on the Physical and Sensory Properties of Lemuru (Sardinella lemuru) Fish Crackers

Fenny Crista Anastasia Panjaitan — 2026-09-27

Fish crackers are a widely consumed ready-to-eat snacks in Southeast Asian. They are rich in carbohydrates and protein, but low levels of dietary fiber and minerals. Recently, Ulva sp. has received increasing attention as a new sustainable food resource due to its high dietary fiber and nutritional value. Its incorporation into food products may enhance food diversification and quality. This study aimed to evaluate the effects of Ulva sp. incorporation on the physical characteristics and sensory properties of lemuru (Sardinella lemuru) fish crackers, as well as to determine the most acceptable formulation for product quality. Results revealed that Ulva sp. supplementation significantly increased product yield (p<0.05), while linear expansion and oil absorption decreased with increasing concentration. After deep-frying, crackers with 2% Ulva sp. showed the highest linear expansion (82.24 ± 3.68%) and oil absorption (12.62 ± 0.90%, p<0.05), whereas those with 10% Ulva sp. showed the lowest levels, at 51.06 ± 7.53% for linear expansion (p<0.05) and 8.38 ± 0.76% for oil absorption (p<0.05). Sensory evaluation indicated that the addition of 10% Ulva sp. significantly (p<0.05) reduced the consumer acceptance of the appearance attribute, while odor, taste and texture were not significantly altered. Based on overall attribute, lemuru cracker with 5% Ulva sp. was identified as the optimal formulation. This study contributes to use of Ulva sp. as a functional ingredient in resulting food products with acceptable product quality.

Application of Pulsed Electric Field (PEF) for Improving the Quality of MSME Tomato Sauce: A Comprehensive Pre- and Post-Treatment Analysis

Meiria Istiana Sari — 2026-09-26

The increasing demand for high-quality, minimally processed foods has driven interest in non-thermal technologies such as pulsed electric field (PEF). This study evaluated the effect of PEF treatment on Deepika tomato sauce produced by micro, small, and medium enterprises (MSME) in Malang, Indonesia. Samples were divided into control and PEF-treated groups. PEF was applied at 10 kV/cm, with pulse frequencies of 1–1000 Hz for approximately 2 seconds. pH, aroma, taste, proximate composition, heavy metals, and microbiological quality were compared before and after treatment. The pH remained 3.8, while the characteristic tomato aroma and taste were retained. Crude fiber changed from 2.09% to 2.08%, ash from 3.73% to 3.75%, carbohydrate from 26.40% to 23.34%, protein from 0.89% to 0.07%, and crude fat from 8.66% to 8.40%. Pb, Cd, Hg, and As were not detected. Cu remained 0.8 mg/kg and Zn changed from 3.7 to 3.3 mg/kg. TPC decreased from 3.4 × 10⁴ to 9.2 × 10² CFU/mL (1.57 log reduction), while YMC decreased from 6.2 × 10⁴ to 3.8 × 10¹ CFU/mL (3.21 log reduction). This study contributes to provide an integrated pre- and post-treatment assessment of PEF applied to an MSME-scale viscous tomato sauce in Indonesia, linking sensory, nutritional, heavy-metal, and microbiological quality within one evaluation framework. The findings support further optimization of PEF parameters for small-scale tomato-sauce processing.

Mathematical Modelling, Drying Kinetics, and Moisture Diffusion Behaviour of Leek Leaves (Allium ampeloprasum var. porrum) during Hot-Air Oven Drying

Satria Bhirawa Anoraga — 2026-06-26

This study investigated the drying kinetics, mathematical modelling, and moisture diffusion behaviour of leek leaves (Allium ampeloprasum var. porrum) during hot-air oven drying at 45, 55, and 65 °C. The study aimed to identify the most suitable thin-layer drying model and determine the effective moisture diffusivity and activation energy of leek leaves. Drying occurred entirely in the falling-rate period, which indicates that internal diffusion controlled moisture removal. Among the evaluated models, the cubic model gave the best fit at 45 °C, with an R² of 98.96%, root mean square error (RMSE) of 0.053, and sum of squared errors (SSE) of 0.0593. The Wang and Singh model gave the best fit at 55 °C and 65 °C, with R² values of 92.26% and 93.17%, RMSE values of 0.078 and 0.102, and SSE values of 0.0874 and 0.1140, respectively. Effective moisture diffusivity increased from 9.13 × 10⁻⁹ m²/s at 45 °C to 31.69 × 10⁻⁹ m²/s at 55 °C and 43.36 × 10⁻⁹ m²/s at 65 °C. The activation energy reached 70.08 kJ/mol, showing the temperature sensitivity of moisture migration in leek tissue. These findings provide useful kinetic parameters for selecting drying conditions and improving predictive modelling in small-scale and industrial leek drying processes. This study contributes part-specific drying data for the green leaf portion of leek, which remains less represented than whole-slice and pseudo-stem drying data.

Automated Multi-Scenario Classification of Food Allergens in Indonesian Culinary Datasets Using Bernoulli Naïve Bayes with Augmentation and Oversampling Methods

Aji Prasetya Wibawa — 2026-06-26

Limited awareness of allergen content in traditional Indonesian foods poses a potential health risk because recipe platforms rarely provide explicit allergen information. This study develops an automated allergen classification system for Indonesian recipes using the Bernoulli Naïve Bayes algorithm. The contributions include the construction of a recipe-level allergen dataset, the integration of text preprocessing, augmentation, and resampling techniques, and the introduction of multi-scenario allergen classification to support safer food choices. Recipe data were collected from Cookpad, producing 9,921 cleaned recipes after preprocessing. To increase data diversity and address class imbalance, rule-based ingredient augmentation expanded the dataset to 15,030 recipes. The model was evaluated under three classification scenarios: binary allergen detection (0–1), allergen exposure levels (0–3), and detailed allergen counts (0–14). Each scenario was tested using four dataset configurations: raw data, raw + ROS (Random Oversampling), augmented data, and augmented + ROS. Class distributions become increasingly imbalanced as the number of classes grows, particularly in the 15-class scenario. The binary scenario achieved perfect performance, with accuracy, precision, recall, and F1-score all equal to 1.0 across all dataset variants. Scenario 2 achieved the best multi-class performance with an F1-score of 0.9293 using Augmented + ROS data. Performance decreased in the 15-class setting due to increased classification complexity. Although a perfect binary score may raise concerns about overfitting, the result likely reflects the clear separability of the dataset’s allergen features. It should be interpreted cautiously when applied to broader culinary data.

Detection Metabolite Volatile Compounds and In Silico Bioactivity of Robusta Leaves (Coffea canephora) from Pakusari Village, Jember

Abdillah Maulana Farhan — 2026-09-26

Robusta coffee is among the most plentiful varieties, although the utilisation of its foliage remains suboptimal. People belief that coffee leaves help reduce hypertension. The objective of this research was to identify volatile metabolites from robusta coffee foliage utilising GC-MS and this study will contribute to advances in the further development of natural phytopharmaceuticals. This study acts as a preliminary evaluation for the creation of herbal treatments derived from coffee leaves. The primary technique of GC-MS involves identifying volatile metabolites using molecular weight, retention time, and percentage area. The research employed maceration extraction over three days with a leaf sample ratio of 30 g to 300 mL of 96% methanol, chemical identification via GC-MS, and in silico bioactivity assessment using PASS Online. The results of GC-MS detection showed that robusta coffee leaf extract contained 73 metabolite compounds. Dominant compounds in coffee leaf extract included 2-Propanol, 1,1'-oxybis- (peak 18), 1-Propanol, 2-(2-hydroxypropoxy)- (peak 21), Butanedioic acid, monomethyl ester (peak 43), Benzofuran, 2,3-dihydro- (peak 46), Succinimide (peak 49), 5-Hydroxymethylfurfural (peak 50), Phenol, 4-ethenyl-2,6-dimethoxy- (peak 52), Phytol (peak 53), Catechol (peak 57), Niacin (peak 58), n-Hexadecanoic acid (peak 61), Hydroquinone (peak 63), 9-Octadecenoic acid, (E)- [Elaidic acid] (peak 65), Caffeiene (peak 67), 9,12-Octadecadienoic acid (Z,Z)- [linoeic acid] (peak 68), Xanthine, 1,3,7,8-tetramethyl- (peak 69), 9,12,15-Octadecatrienoic acid, (Z,Z,Z)- (peak 70), dan 1,3-Propanediol, 2-(hydroxymethyl)-2-nitro- (peak 71). The area percentage value of caffeine was 45.37%, and in silico bioactivity showed analeptic characteristics. This analysis could not match the phenomenon in Pakusari Village, where coffee leaves are antihypertensive.

Effects of Pollen Layer Thickness and Botanical Origin on Color and Moisture Content in Dried Bee Pollen

Ulfah Anis — 2026-09-26

Bee pollen is used as a substitute ingredient, a fortification material, and a dietary supplement. It contains both macro- and micronutrients such as carbohydrates, proteins, lipids, minerals, and vitamins. In addition, bee pollen has a role as an antioxidant and antibacterial. Its chemical and physical characteristics differ by several factors, particularly its botanical origins. Bee pollen is often incorporated into food products in dried form, which offers a longer shelf life, compared to fresh bee pollen. The drying method is a factor that affects the quality of bee pollen. While previous research has focused primarily on drying temperatures, this study differentiates itself by investigating how the interaction between pollen layer thickness and botanical origin affects color and moisture content. This reveals how different floral species respond to physical heat-and-mass limitations during the drying process. The contribution of this research was to investigate the effect of layer thickness during drying in the container and botanical origin, on the color and moisture content of the dried bee pollen. The research experimental design was a completely randomized design (CRD) with two factors: thickness (0 cm; 1 cm; 1.5 cm) and botanical origins. The experiment consisted of 9 treatments, each executed with 3 repetitions. Data were analyzed using the Analysis of Variance (ANOVA); if there was a significant effect, the Duncan Multiple Range Test (DMRT) was conducted with α=5%. The results showed that the interaction between layer thickness and botanical origins of bee pollen significantly affected a^* value, and hue angle. On the contrary, the interaction between the two factors had no significant effect on the moisture content, chroma, and b^* value. Dried bee pollen had moisture content of just under 10%. Quantitatively, the applied drying conditions successfully reduced the moisture content below 10%, while maintaining a lightness (L^*) of 31–49, an a^* value of 4–16, a b^* value of 20–30, a hue angle of 58–82 and a chroma 20–30.