İlgi alanları

Sürdürülebilir Kalkınma Amaçları

• 5
• 1
• 1

• MakaleYalnız künye2026Scopus SJRQ2/r/bayburt/11378
  Leakage-safe benchmarking of conventional chemometric and deep learning models for food NIR spectroscopy regression

  Akpolat, Hacer; Akpolat, Mehmet Zahid

  Near-infrared (NIR) spectroscopy is widely used in food systems for rapid, non-destructive estimation of compositional and quality attributes. Although machine-learning and deep-learning models are increasingly reported for spectral regression, practical model selection is often hindered by heterogeneous validation setups and the risk of information leakage during preprocessing and tuning. In this study, we provide a leakage-safe benchmark across nine food-related regression tasks spanning different sample sizes and spectral dimensionalities. Conventional chemometric and regression baselines (partial least squares regression, ridge regression, and support vector regression) are compared with lightweight deep-learning baselines (a shallow artificial neural network and a one-dimensional convolutional neural network) under a unified leakage-safe evaluation framework combining nested cross-validation with acomplementary fixed held-out test-set assessment, fold-wise preprocessing, and repeated training for stochastic models. Performance is summarized primarily using prediction error (RMSE in nested cross-validation and RMSEP on the held-out test set), while the coefficient of determination is retained as a secondary descriptive measure, complemented by cross-validation and fixed held-out-test learning-curve analyses together with model-complexity analysis. Across both validation analyses, ridge regression achieved the largest number of numerical task-level wins, while partial least squares regression and support vector regression remained competitive across multiple tasks. Paired randomization tests following van der Voet further showed that some small numerical differences between models were not statistically distinguishable. Artificial neural network and one-dimensional convolutional neural network baselines did not provide a reliable generalization advantage and showed greater variability in some data-limited settings. From a food-measurement perspective, these findings indicate that well-validated conventional models remain a strong default for routine NIR calibration workflows and non-destructive quality assessment. Deep learning appears most justified when calibration settings provide larger datasets, stronger experimental control, or spectroscopy-specific inductive bias that can translate added model complexity into practical measurement gains. Overall, this study serves as a controlled benchmark and reference for model selection under realistic food-spectroscopy data regimes.

  • 
  AtıfOpenAlex1

  İncele
• Kitap BölümüAçık Erişim2024/r/bayburt/13063
  Raman Spectroscopy for Food Quality: Assessment Uncovering Adulteration and Ensuring Authenticity

  Ayvaz, Huseyin; Yao, Siyu; Morta, Mustafa; Temizkan, Riza; Akpolat, Hacer; Rodriguez-Saona, Luis E.

  The globalization of the food market, the prevalence of economically motivated adulteration, and the scarcity of high-value food sources have posed significant challenges to ensuring food authenticity in the industry. The assessment of food authenticity has now become a matter of utmost importance, necessitating the adoption of rapid and robust methods. While well-established techniques are available for analysis, many are not always feasible because of the need for high-priced instrumentation and proficient operators, excessive time requirements, and the destructive nature of the methods. However, Raman spectroscopy has been positioned as an attractive technology offering high-speed and nondestructive fingerprinting capabilities to monitor sample characteristics of foods. Improvements in optical technology and portable instruments have allowed field-deployable devices and have made it practical to detect adulteration from production to the market. An extensive literature has demonstrated the convenience and efficacy of these instruments in fulfilling the needs of the food industry. This chapter highlights the recent advancements in research studies focusing on the utilization of Raman spectroscopy for the rapid determination of food adulteration and authentication. These developments showcase the potential of Raman spectroscopy as a valuable tool in addressing the challenges posed by food fraud. The ongoing progress in Raman spectroscopy and its applications in food authenticity assessment represents a significant step forward in combating adulteration and ensuring the integrity of our food supply. Further research and technological advancements in this field will enhance the capabilities of Raman spectroscopy, providing the industry with increasingly reliable, efficient, and practical methods to combat food adulteration and authenticate the foods we consume.

  • 
  İncele
• MakaleAçık Erişim2023/r/bayburt/15675
  The impact of the COVID-19 pandemic on eating and food shopping habits

  Demirer, Büşra; Akpolat, Hacer; Bayraktar, Mukaddes Kılıç

  Objective: This study aimed to investigate the effect of the COVID-19 pandemic on eating and food shopping habits among the Turkish adult population. Material and method: Demographics, eating and food shopping habits, and food label reading habits of the participants were collected via online surveys. Coronavirus anxiety was assessed using the Coronavirus Anxiety Scale. The survey was conducted from November 2021 to the end of January 2022. Student's t-test was used to determine the statistical difference between quantitative variables. Chi-Square and Marginal Homogeneity Tests, depending on the number of categories, were used to determine the difference between qualitative variables. Results and discussion: Unpackaged food consumption decreased during the pandemic. More than half of the participants started to pay more attention to food labels, spend less time for grocery shopping, and started using nutritional supplements. The changes in eating, grocery shopping, and food label reading habits among Turkish consumers during the pandemic have been demonstrated.

  • 
  AtıfOpenAlex3

  İncele
• MakaleAçık Erişim2020/r/bayburt/14779
  Determination of Optimum Ethanolic Extraction Conditions and Phenolic Profiles of Thyme, Mint, Uckun, Grape Seeds and Green Tea Waste Fiber

  Akpolat, Hacer; Türkhan, Ayşe; Tunçtürk, Yusuf; Bulu, Menekşe; Alwazeer, Duried

  The objective of this study was to investigate the effect of different ethanol ratios inextraction solvent as well as the antioxidant properties of five plants. Thyme, mint, uckun, grapeseeds and green tea waste fiber was analyzed to determine total phenolic content (TPC) andantioxidant activity by ABTS and DPPH radical scavenging activity assays. Individual phenoliccomponents were analyzed with reverse phase high performance liquid chromatography (HPLC).TPC varied significantly from 2.00±0.27 to 172.68±0.19 mg GAE g-1 dw depending on the plant typeand ethanol ratio of the solvent. The effect of ethanol ratio also varied among different plants. HPLCanalysis was performed for the extracts showing highest antioxidant activity, and green tea wastefiber (699.89 mg 100 g-1 dw) had the highest concentration of phenolic compounds overall, whilemint (173.67 mg 100 g-1 dw) had the lowest amount. Correlations between TPC and antioxidantactivity was significant which is comparable to the previous report.

  • 
  AtıfOpenAlex12·TR Dizin1

  İncele
• MakaleAçık Erişim2020Scopus SJRQ1/r/bayburt/13176
  Portable infrared sensing technology for phenotyping chemical traits in fresh market tomatoes

  Akpolat, Hacer; Barineau, Mark; Jackson, Keith A.; Aykas, Didem P.; Rodriguez-Saona, Luis E.

  Our objective was to develop predictive regression algorithms based on infrared spectroscopy to screen for selected quality traits directed at optimizing the selection capabilities of fresh market tomatoes. Fresh tomato (681) samples were harvested from multiple locations (Florida, Virginia, California and South Carolina) during the 2016 and 2018 seasons at various ripening stages. Spectra were collected by transmittance and attenuated total reflectance (ATR) either from the tomato surface or juice. Reference methods included soluble solid content, titratable acidity, sugars, organic acids, and lycopene. Partial least squares regression using surface spectra showed good correlation for lycopene and ascorbic acid (rcv>0.9) but modest correlation coefficients (rcv 0.54–0.80) for all other traits, while juice spectra gave high correlation coefficients (rcv>0.94) and excellent predictive performance (RPD range 3–10) for all quality traits except ascorbic acid (rcv>0.79). Multiple quality traits were simultaneously determined by using a single drop of sample providing fast (<1 min) measurements and minimal sample preparation based on unique spectral fingerprints.

  • 
  AtıfScopus17·WoS15·OpenAlex19

  İncele
• MakaleAçık Erişim2020Scopus SJRQ2/r/bayburt/12033
  High-throughput phenotyping approach for screening major carotenoids of tomato by handheld raman spectroscopy using chemometric methods

  Akpolat, Hacer; Barineau, Mark; Jackson, Keith A.; Akpolat, Mehmet Z.; Francis, David M.; Chen, Yu-Ju; Rodriguez-Saona, Luis E.

  Our objective was to develop a rapid technique for the non-invasive profiling and quantification of major tomato carotenoids using handheld Raman spectroscopy combined with pattern recognition techniques. A total of 106 samples with varying carotenoid profiles were provided by the Ohio State University Tomato Breeding and Genetics program and Lipman Family Farms (Naples, FL, USA). Non-destructive measurement from the surface of tomatoes was performed by a handheld Raman spectrometer equipped with a 1064 nm excitation laser, and data analysis was performed using soft independent modelling of class analogy (SIMCA)), artificial neural network (ANN), and partial least squares regression (PLSR) for classification and quantification purposes. High-performance liquid chromatography (HPLC) and UV/visible spectrophotometry were used for profiling and quantification of major carotenoids. Seven groups were identified based on their carotenoid profile, and supervised classification by SIMCA and ANN clustered samples with 93% and 100% accuracy based on a validation test data, respectively. All-trans-lycopene and β-carotene levels were measured with a UV-visible spectrophotometer, and prediction models were developed using PLSR and ANN. Regression models developed with Raman spectra provided excellent prediction performance by ANN (rpre = 0.9, SEP = 1.1 mg/100 g) and PLSR (rpre = 0.87, SEP = 2.4 mg/100 g) for non-invasive determination of all-trans-lycopene in fruits. Although the number of samples were limited for β-carotene quantification, PLSR modeling showed promising results (rcv = 0.99, SECV = 0.28 mg/100 g). Non-destructive evaluation of tomato carotenoids can be useful for tomato breeders as a simple and rapid tool for developing new varieties with novel profiles and for separating orange varieties with distinct carotenoids (high in β-carotene and high in cis-lycopene).

  • 
  AtıfScopus40·WoS32·OpenAlex48

  İncele
• MakaleAçık Erişim2015Scopus SJRQ1/r/bayburt/14104
  The Effect of pH and Temperature on Cabbage Volatiles During Storage

  Akpolat, Hacer; Barringer, Sheryl Ann

  During storage of shredded cabbage, characteristic sulfurous volatile compounds are formed affecting cabbage aroma both negatively and positively. Selected ion flow tube-mass spectrometry (SIFT-MS) was used to measure the concentration of cabbage volatiles during storage. The volatile levels of cabbage samples were measured at pH 3.3 to 7.4 at 4 degrees C for 14d, and pH 3.3 at 25 degrees C for 5d in order to determine the effect of pH and temperature. Aroma intensity, best aroma, freshness, and off odor were evaluated in a sensory test of the samples at 4 degrees C. The desirable volatile allyl isothiocyanate was lower in high pH samples (pH 7.4 and 6.4), whereas higher concentrations were detected in low pH samples (pH 3.3 and 4.6). Lipoxygenase volatiles, which produce a fresh green and leafy aroma in cabbage, were generated in very low amounts at any pH value. High pH samples generated significantly higher concentrations of off odors such as dimethyl sulfide, dimethyl disulfide, dimethyl trisulfide, and methanethiol. Sensory tests showed that higher pH samples had significantly stronger off odor and lower desirable cabbage aroma than lower pH samples. Thus, sensory results matched the volatile results in that samples at higher pH levels formed the highest amount of undesirable volatiles and the least amount of desirable volatiles. Storage at 25 degrees C produced similar concentrations of allyl isothiocyanate, but significantly higher levels of off odors, than at 4 degrees C. Shredded cabbage products should be stored in low pH dressings to minimize formation of off odors and maximize formation of characteristic, desirable cabbage odor. Practical Application Low pH samples formed fewer undesirable, and more desirable volatile compounds overall. The natural pH of cabbage is around 6.4, and at this pH, higher concentrations of off odors are formed during storage. Thus, low pH salad dressings are more suitable for storage of ready-to-eat cabbage products in terms of sensory quality, and high temperature is not suitable for cabbage storage, since undesirable aroma compound formation is high, desirable aroma loss occurs quickly, and microbial growth occurs at a very early stage of storage.

  • 
  AtıfWoS27·OpenAlex32

  İncele