Multivariate regression methods with infrared spectroscopy to detect the falsification of traditional butter

Abstract

Recently, food safety and guaranteed of food marks have become more important subjects of foodstuff production and the marketing of processed foods. This paper demonstrates the ability of Mid Infrared spectroscopy coupled with multivariate regression tools to detect vegetable butter (as adulterant) in a binary mixture with traditional cow’s butter. Blends of traditional cow’s butter with different percentages of vegetable butter were measured using Attenuated Total Reflectance-Fourier Transform Mid Infrared Spectroscopy (ATRFTMIR). Spectral and reference data were firstly analyzed by principal component analysis (PCA) to check outliers samples; and improve the robustness of the prediction models to be established. Multivariate regression methods as Principal component regression (PCR) and Partial least square regression (PLSR) were used to establish calibration model. Excellent correlation between ATR-FTMIR analysis and studied butter blends was obtained R2 = 0.99; with Root Mean Square Errors of Prediction < 3.04, Limit of Detection 9.12% (By PCR) and 6.06% (by PLSR), and Relative Prediction Errors as low as 3.13.

Authors and Affiliations

W. Terouzi

Keywords

Related Articles

Electronic Customer Relationship Management and Consumer Behaviour (A Study of Diamond Bank Plc, Nigeria)

The study focuses on electronic customer relationship management and consumer behavior with special emphasis on diamond bank Plc. Retail financial services in all markets, including emerging markets, are undergoing major...

Food Safety: A Primer

Public health is directly or indirectly affected by the food supply. Food safety is a public health issue which has become a global problem. Since food is necessary for human survival, a food safety crisis can cause wide...

Information Security Adherence in Institutions of Higher Education: A Case Study of American University of Nigeria

Institutions of higher education face a higher level of information security incidence and compromise such as data and information theft, malicious program infection, attack on the information systems infrastructure and...

A Brief History of Computation's in Portugal

A history of the great development of information technology in the twentieth century, at a global level. The story of the emergence and development of information technology in Portugal, also in the twentieth century, i...

Influences of Brand Image on Brand Equity and Brand Loyalty in case of Mongolian Banking Sector

In 21th century’s competitive business environment, commercial organizations need to get advantage compared to competitors in order to survive. Because customer intention is changed always by result of competition. In co...

Download PDF file
  • EP ID EP363568
  • DOI -
  • Views 134
  • Downloads 0

How To Cite

W. Terouzi (2018). Multivariate regression methods with infrared spectroscopy to detect the falsification of traditional butter. Invention Journal of Research Technology in Engineering & Management, 2(7), 87-96. https://europub.co.uk./articles/-A-363568