Scholarlink

<p>Fortification of yogurt with zinc oxide (ZnO) offers a promising approach to enhance its nutritional and structural properties. This study examined the effects of micro- and nano-sized ZnO particles on the rheological and textural behavior of cow milk yogurt and explored principal component analysis (PCA) and machine learning for multivariate interpretation and prediction. Yogurts containing 75&#xa0;ppm micro-ZnO, 75&#xa0;ppm nano-ZnO, and a control were evaluated for yield stress, plastic viscosity, consistency coefficient, and flow behaviour index, with the Power Law model providing the best fit for flow data. Firmness and cohesiveness were monitored during 7&#xa0;days of refrigerated storage (at 7 ± 1&#xa0;°C). Nano-ZnO yogurt exhibited the highest yield stress (13.23 ± 0.01&#xa0;Pa) and consistency coefficient (7.720&#xa0;Pa·sⁿ), along with the lowest plastic viscosity (0.0949&#xa0;Pa-s) and flow behaviour index (0.224), confirming its superior structural rigidity and pronounced shear-thinning characteristics. PCA revealed clear clustering among treatments, driven by differences in rheological–textural attributes. Machine learning models accurately predicted firmness, with K-Nearest Neighbours achieving the highest performance (R<sup>2</sup> = 0.857), while classification algorithms effectively distinguished yogurt types. Overall, ZnO nanoparticles substantially enhance yogurt gel strength and stability, and the integrated PCA–ML approach provides a robust framework for quality prediction and product differentiation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine learning approaches for rheology and texture profiling of yogurt fortified with nano and microscale zinc oxide

  • 期刊论文

  • 发表日期：

    2026年6月19日

查看全文
  • Payal Karmakar,
  • Pinaki Ranjan Ray,
  • Ankita Pandit,
  • Pareshnath Chatterjee,
  • Ronit Mandal

摘要

Fortification of yogurt with zinc oxide (ZnO) offers a promising approach to enhance its nutritional and structural properties. This study examined the effects of micro- and nano-sized ZnO particles on the rheological and textural behavior of cow milk yogurt and explored principal component analysis (PCA) and machine learning for multivariate interpretation and prediction. Yogurts containing 75 ppm micro-ZnO, 75 ppm nano-ZnO, and a control were evaluated for yield stress, plastic viscosity, consistency coefficient, and flow behaviour index, with the Power Law model providing the best fit for flow data. Firmness and cohesiveness were monitored during 7 days of refrigerated storage (at 7 ± 1 °C). Nano-ZnO yogurt exhibited the highest yield stress (13.23 ± 0.01 Pa) and consistency coefficient (7.720 Pa·sⁿ), along with the lowest plastic viscosity (0.0949 Pa-s) and flow behaviour index (0.224), confirming its superior structural rigidity and pronounced shear-thinning characteristics. PCA revealed clear clustering among treatments, driven by differences in rheological–textural attributes. Machine learning models accurately predicted firmness, with K-Nearest Neighbours achieving the highest performance (R2 = 0.857), while classification algorithms effectively distinguished yogurt types. Overall, ZnO nanoparticles substantially enhance yogurt gel strength and stability, and the integrated PCA–ML approach provides a robust framework for quality prediction and product differentiation.