School of Industrial Technology, Environmental Technology Division, Universiti Sains Malaysia, Penang, Malaysia.
Int J Food Sci Nutr. 2009;60 Suppl 4:116-25. doi: 10.1080/09637480802609368.
Major (sodium, potassium, calcium, magnesium) and minor elements (iron, copper, zinc, manganese) and one heavy metal (lead) of Cavendish banana flour and Dream banana flour were determined, and data were analyzed using multivariate statistical techniques of factor analysis and discriminant analysis. Factor analysis yielded four factors explaining more than 81% of the total variance: the first factor explained 28.73%, comprising magnesium, sodium, and iron; the second factor explained 21.47%, comprising only manganese and copper; the third factor explained 15.66%, comprising zinc and lead; while the fourth factor explained 15.50%, comprising potassium. Discriminant analysis showed that magnesium and sodium exhibited a strong contribution in discriminating the two types of banana flour, affording 100% correct assignation. This study presents the usefulness of multivariate statistical techniques for analysis and interpretation of complex mineral content data from banana flour of different varieties.
卡文迪什香蕉粉和梦想香蕉粉中的主要（钠、钾、钙、镁）和微量（铁、铜、锌、锰）元素以及一种重金属（铅）进行了测定，并采用因子分析和判别分析等多元统计技术对数据进行了分析。因子分析产生了四个解释超过 81%总方差的因子：第一个因子解释了 28.73%，包括镁、钠和铁；第二个因子解释了 21.47%，仅包含锰和铜；第三个因子解释了 15.66%，包含锌和铅；第四个因子解释了 15.50%，包含钾。判别分析表明，镁和钠在区分两种香蕉粉方面具有很强的贡献，能够实现 100%正确的归属。本研究展示了多元统计技术在分析和解释不同品种香蕉粉复杂矿物质含量数据方面的有用性。
