Original Article Temperature compensation for determination of moisture and reducing sugar of longan honey by near infrared spectroscopy Chalermpun Thamasopinkul 1 , Pitiporn Ritthiruangdej 1 , Sumaporn Kasemsumran 2 , Thongchai Suwonsichon 1 , Vichai Haruthaithanasan 1 and Yukihiro Ozaki 3 Abstract Nearinfrared spectra of honeys are affected by sample temperature variation, mainly due to a change in hydrogen bonding of water. The aim of this study was to develop robust and powerful calibration models which can compensate for a variation of sample temperature for the determination of moisture and reducing sugar content in honey using near infrared spec- troscopy. Partial least squares regression with the aid of standard normal variate transformation was used to develop three calibration models at constant temperature (25, 35 and 45  C) and a robust calibration model with temperature compen- sation. All the developed models for moisture and reducing sugar content showed high performance of prediction with coefficient of determination (r 2 ) and residual prediction deviation values greater than 0.95 and 3.8, respectively. The results show that the temperature compensation model can be considered as a robust calibration model for near infrared deter- mination of moisture and reducing sugar in the honey when sample temperature is varied. Keywords Honey, temperature variation, temperature compensation, near infrared spectroscopy, partial least squares regression, non- destructive, standard normal variate Received 3 September 2016; accepted 5 November 2016 Introduction Honey is a viscous, aromatic, sweet food that consists mainly of sugars (80–85%), water (15–21%), and other substances such as proteins, organic acids, vitamins, minerals, pigments, phenolic compounds, a large var- iety of volatile compounds, and solid particles derived from honey harvesting. Monosaccharides represent about 75% of the sugars found in honey, along with 10–15% disaccharides and small amounts of other sugars. 1 Generally, the sugars and water content pre- sent in honeys depend mainly on the botanical origin of the honey, the level of maturity achieved in the hive, the processing techniques and storage conditions. 2 The purpose of laws regarding honey is to ensure the identity and minimum quality requirements for honey. The amount of sugars and water are very important indicators in evaluating the maturity of honey. Moisture content of honey is a limiting factor in the determination of its physical properties (viscosity and crystallisation, specific gravity, solubility and conserva- tion), sensory properties (colour, flavour and taste) and spoilage resistance against yeast fermentation. 3 The higher the moisture content, the higher is the probabil- ity of undesirable honey fermentation during storage caused by the action of osmotolerant yeasts. 4 Almost all standard analytical techniques for deter- mining the chemical constituents of honey are time consuming, expensive, and they require specialised knowledge and expertise and involve laborious 1 Faculty of Agro-Industry, Department of Product Development, Nondestructive Quality Analysis (NQA) Center, Kasetsart University, Bangkok, Thailand 2 Kasetsart Agricultural and Agro-Industrial Product Improvement Institute (KAPI), Kasetsart University, Bangkok, Thailand 3 Department of Chemistry, School of Science and Technology, Kwansei Gakuin University, Hyogo, Japan Corresponding author: Pitiporn Ritthiruangdej, Faculty of Agro-Industry, Department of Product Development, Nondestructive Quality Analysis (NQA) Center, Kasetsart University, 50 Chatuchak, Bangkok 10900, Thailand. Email: pitiporn@gmail.com Journal of Near Infrared Spectroscopy 2017, Vol. 25(1) 36–44 ! The Author(s) 2017 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0967033516685549 journals.sagepub.com/home/jns