Calibration Set Research Articles

Identification of fermented soy sauce and blended soy sauce based on dielectric spectra

A new method based on dielectric properties has been developed to distinguish blended soy sauce from fermented soy sauce. Fifty samples of pure fermented soy sauce and blended soy sauce with different contents of hydrolyzed vegetable protein (HVP) were prepared. The dielectric constant and dielectric loss factor of all soy sauce samples in the 30 MHz–3000 MHz frequency range were measured using an impedance analyzer. The sample set was divided into a correction set and a prediction set using the joint x-y distances (SPXY) algorithm, and the partial least squares (PLS) and support vector machine (SVM) models were adopted to distinguish the different samples. The effects of selecting characteristic variables on model prediction using the full spectra (FS), principal component analysis (PCA), and successive projection algorithm (SPA) were compared. Results indicate that the discriminant effect of the PLS model was better than that of the SVM model overall. The SPA–PLS model had the best predictive performance among the six developmental models. The correlation coefficients of the correction set and prediction set were 0.9205 and 0.9096, respectively. The root mean square error of the calibration set was 1.3699 and that of the prediction set was 1.5950. The study demonstrated that the combination of dielectric spectra and stoichiometry can be utilized to determine whether soy sauce has been adulterated.

90Y SIR-Spheres Activity Measurement with New SIROS D-Vial Delivery Kit.

A new 90Y SIR-Spheres delivery kit (SIROS D-vial and shield) has been introduced with a different physical form from the legacy V-Vial kit. Here, we establish the dose calibrator settings and exposure-rate-to-activity conversion factor to assay 90Y SIR-Spheres activity in the new SIROS kit. Methods: Eight D-vials with initial 90Y activities from 1.2 to 6.6 GBq within acrylic shields were assayed with dose calibrators and exposure-rate meters until activities decayed to approximately 0.1 GBq. The dose calibrator settings resulting in the lowest median activity errors and the best-fit slope of exposure rate versus activity were identified. Results: SIROS D-vial 90Y activity can be accurately and reliably estimated directly using setting 51 × 10 on both the CRC-15R and the CRC-55tR dose calibrators (errors within ±0.5%) and indirectly with an exposure-rate reading at 30 cm using conversion factor 0.664 ± 0.003 GBq/(mR/h) (R 2 = 0.985). Conclusion: Dose calibrator settings and exposure-rate-to-activity conversion factor for 90Y activity assays with new SIROS kit should be updated from legacy V-Vial parameters to avoid an approximately 10% underestimation.

Quantitative determination of oxides in cement raw meal based on near-infrared spectroscopy and hybrid feature selection strategy

Determining the content of components in cement raw meal is vital in Portland cement manufacturing. Near-infrared spectroscopy (NIRS)enables rapid, safe, and accurate quantitative analysis of the primary constituents of cement raw meal (CaO, SiO2, Al2O3, and Fe2O3), thereby assisting the cement industry in reducing the time delay in raw meal proportioning control and enhancing the quality of the raw meal. This study employed a hybrid feature selection strategy to identify effective feature points in the NIRS of raw meal samples. Effective feature points from the raw meal were integrated with partial least squares (PLS) to develop a regression model. Initially, NIRS data from raw meal samples were obtained using a Fourier NIR spectrometer. Subsequently, sample set partitioning based on joint X-Y distance (SPXY) was utilized to divide the samples into a calibration set (70 samples) and a validation set (22 samples). Next, A backward interval partial least squares (BiPLS) approach, along with various preprocessing algorithms, was then employed to initially screen the optimal waveband combinations of the NIRS. Within these combinations, effective feature points were further identified using competitive adaptive reweighted sampling (CARS), variable combined population analysis (VCPA) and weighted fusion variable space shrinkage approach (WFVSSA). Finally, we developed and compared PLS models based on the full spectrum and three hybrid feature selection methods. The results demonstrated that.WFVSSA offers a comprehensive assessment of wavelength importance, effectively preserving feature points while eliminating redundant information. BiPLS-WFVSSA-PLS outperformed BiPLS-CARS-PLS and BiPLS-VCPAPLS on the validation set, with R2, RMSEP, and RPD values of 0.8676, 0.1431, and 2.8129 for CaO; 0.9212, 0.1216, and 3.6461 for SiO2; 0.9198, 0.0607, and 3.5618 for Al2O3; and 0.8717, 0.0166, and 2.8614 for Fe2O3, respectively. These findings underscore the superior accuracy and stability of BiPLS-WFVSSA-PLS in determining cement raw meal component contents.

Rapid classification of rice according to storage duration via near-infrared spectroscopy and machine learning

Rice is the most important staple crop for more than half of the world's population. As rice quality can deteriorate during storage, methods that can effectively classify rice according to its storage duration are essential. However, existing methods of assessing rice storage time are time-consuming, laborious, and incompatible with modern industrial processing technologies. Therefore, we investigated the ability of near-infrared spectroscopy combined with machine learning algorithms to distinguish rice storage duration. A total of 482 rice samples were analyzed, which included 74, 100, and 308 samples produced during 2015–2016, 2017–2018, and 2020–2021, respectively. Five pre-processing methods were initially applied to the spectra to enhance the accuracy of the discrimination model. Subsequently, two-dimensional correlation spectroscopy and competitive adaptive reweighted sampling (CARS) were used to extract the characteristic spectra associated with storage time. Finally, three pattern recognition methods (K-nearest neighbor analysis, linear discriminant analysis, and least squares support vector machine (LS-SVM)) were compared for their effectiveness in constructing classification models. The results indicated that the best model for identifying the storage duration of rice was established after spectral pre-processing with the standard normal variate and first derivative, using the CARS algorithm to select feature wavelengths, and applying the LS-SVM modeling method, which together yielded correct identification rates of 99.72 % and 91.67 % for the calibration and validation sets, respectively. Thus, we propose near-infrared spectroscopy coupled with machine learning algorithms as an effective approach for classifying rice according to storage duration, which can facilitate evaluations of rice freshness in the market.

Detection of early collision and compression bruises for pears based on hyperspectral imaging technology

Early detection of bruising is one of the major challenges in postharvest quality sorting processes for pears. In this study, visible/near infrared (VIS/NIR) hyperspectral imaging (400–1000 nm) was utilized for early detection of pear bruise type and timing (1, 12, and 24 h post-bruise). Spectral images of nonbruised and mechanically bruised pears (collision and compression) were captured at these intervals for modeling. Spectral data was processed using principal component analysis (PCA) and uninformative variable elimination (UVE) to select optimum wavelengths. Classification models were then built using an extreme learning machine (ELM) and support vector machine (SVM), and compared with a model combining genetic algorithm, sooty tern optimization algorithm, and SVM (STOA-GA-SVM). For PCA-ELM, UVE-ELM, PCA-SVM, and UVE-SVM models, the calibration set accuracies were 98.99%, 98.98%, 96.94%, and 99.23% respectively. And the validation set accuracies were 89.29%, 87.97%, 88.78%, and 88.78% respectively. The STOA-GA-SVM model shows the best performance, and the accuracy of the calibration set and validation set is determined to be 97.19% and 92.86%, respectively. This study shows that the use of the VIS/NIR hyperspectral imaging technique combined with the STOA-GA-SVM algorithm is feasible for the rapid and nondestructive identification of the bruise type and time for pears.

Future changes in soil salinization across Central Asia under CMIP6 forcing scenarios

AbstractSoil salinization is a critical environmental and socio‐economic concern with global implications, and its severity is expected to amplify under changing climate. The impact of climate change on salinization in Central Asia is still not fully understood. This study addresses this gap by employing a digital soil mapping (DSM) framework. Cubist, random forest (RF), and quantile regression forests (QRF) are utilized to project variations in soil surface salinity (0‐10 cm) in Central Asia from 2025 to 2100 under two shared socio‐economic pathways (SSPs): SSP2‐4.5 and SSP5‐8.5. These models are developed using data from 20 global climate models (GCMs) obtained from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The results reveal that the RF model exhibits superior predictive capability in estimating soil salinity. RF performed on the calibration set with a coefficient of determination (R2) of 0.86, root mean square error (RMSE) of 9.84 and 9.90 dS m−1, ratio of performance to interquartile distance (RPIQ) of 3.09 and 3.07, and a Nash–Sutcliffe efficiency (NSE) of 0.86. The multi‐GCM ensemble means revealed the potential for varying degrees of salinization in Central Asia, with higher levels predominantly observed in the southeast and southwest of the study area, particularly downstream of the river and in the lakeside areas. Temporal analysis of soil salinity evolution reveals an overall increase in salinity across the region, with more notable changes projected under SSP5‐8.5. Specifically, the projected increase rate in soil salinity for Central Asia was 0.0005 dS m−1/year under SSP2‐4.5 and 0.01 dS m−1/year under SSP5‐8.5. Turkmenistan is notable for possessing the highest regional average of soil salinity, with the exception of a declining trend observed within this area. The remaining regions of Central Asia exhibit an upward trend in average soil salinity, particularly noteworthy under the SSP5‐8.5 scenario, where variations in soil salinity are more obvious. These findings hold significant potential in enhancing our understanding of how Central Asia responds to global change, advances toward sustainable development, and enhances comprehension of the dynamics within the region.

Size stable batch mode model updating method

In near-infrared spectroscopy analysis, the predictive performance of models may degrade due to variations in the measurement environment and sample properties. Maintenance or updates to the model are essential to uphold prediction accuracy. The state-of-the-art approach for model updates involves labeling new samples and incorporating them into the calibration set. However, this strategy leads to an escalating size of the calibration dataset, and the low ratio of new samples renders the model update process inefficient. To enhance update efficiency, a size-stable updating strategy is presented which involves the simultaneous addition of new samples and removal of old samples. This is achieved by utilizing the new set as a basis for identifying and deleting significantly different old labeled samples. To improve labeling efficiency, a batch mode update is employed. Initially, calibration samples are clustered to yield multiple clusters with comparable sizes to the new update set. Subsequently, differences between multiple old sample clusters and the new set are calculated. The old sample cluster that most different are removed, accompanied by the addition of the new sample set to maintain calibration set size stability. In calculating set similarity, a multi-indices fusion method is employed to ensure accurate judgments. The efficacy of this method is validated through simulated and real data.

Rapid selection of milk mid-infrared spectra for creating a world representative spectral database of dairy cow population

The advantage of employing mid-infrared spectrometry for milk analysis in breeding lies in its ability to quickly generate millions of records. However, these records may be biased if the calibration process does not account for their spectral variability when constructing the predictive model. So, this study introduces a novel method for developing a World Representative Spectral Database (WRSD) to reduce the risks of spectral extrapolation when predicting dairy traits in new samples. Utilizing a 2-phase selection procedure that is both efficient and minimizes memory usage, we first generate a decomposition matrix via Principal Component Analysis (PCA) on a data set of 2,324,443 records. The next phase iterates spectral selection based on a location index from PCA scores, calculating spectra occurrence frequency for refined barycenter estimations. The chosen spectra's barycenter closely aligns with the entire data set, proving the efficacy of using just 3 principal components (PCs). Applied to 4 varied data sets, totaling over 21 million records, we select 583,440 spectra to represent spectral diversity, with selection rates between 2.00% and 14.88%. This selection illustrates the spectral variability across different dairy populations and data providers. Demonstrated through a hypothetical calibration set of 71 samples, the WRSD's utility for algorithm developers becomes apparent. This calibration set covers between 91.42 to 98.50% of the WRSD variability, except for the Irish data set (3.50%), indicating a need for additional samples to accurately represent Irish variability and minimize spectral extrapolation. This study offers valuable insights into the representativeness of training sets for capturing spectral variability within targeted dairy populations. While the current WRSD does not fully encompass global milk spectral diversity, its development underscores the importance of gathering more data and standardizing spectral information across spectrometer brands. Ultimately, the WRSD proves crucial not just for trait prediction but also for identifying abnormal milk samples, also marking a significant relevance in dairy science technology.

Quantitative and qualitative detection of target heavy metals using anti-interference colorimetric sensor Array combined with near-infrared spectroscopy

An anti-interference colorimetric sensor array (CSA) technique was developed for the qualitative and quantitative detection of target heavy metals in corn oil. This method involves a binding mechanism that triggers changes in atomic energy levels and visible color changes. A custom-built olfactory visualization device was employed to gather spectral data, revealing distinct CSA color difference patterns. Subsequently, three pattern recognition algorithms were used to create an identification model for the target heavy metals. The results showed that the ACO-KNN (Ant Colony Optimization-K-Nearest Neighbor) model outperformed the other models, achieving accuracy rates of 90.28% and 89.58% for the calibration and prediction sets, respectively. The ACO-PLS (Partial Least Square) model was more stable with the lowest root mean square error of prediction (RMSEP), which were 0.1730 and 0.1180, respectively. The limit of detection (LOD) and quantification (LOQ) of Pb and Hg were (0.3, 0.6, 1.1 and 2.2) x 10−3 mg/L, respectively.

Low Effectiveness of Mid-Infrared Spectroscopy Prediction Models of Mediterranean Italian Buffalo Bulk Milk Coagulation Traits.

This study evaluated the potential use of mid-infrared spectroscopy to predict milk coagulation traits in bulk milk from Mediterranean Italian buffaloes. A total of 1736 bulk milk samples from 55 farms in central Italy were collected during the official milk quality testing system. The prediction models were developed based on modified partial least-squares regression with 75% of the samples and validated with the remaining samples. All bulk milk samples coagulated between 7.37 and 29.45 min. Average values for milk coagulation traits in the calibration set were 17.71 min, 3.29 min, and 38.83 mm for rennet coagulation time, curd firming time, and curd firmness, respectively. The validation set included samples with similar mean and standard deviation for each trait. The prediction models showed the greatest coefficient of determination of external validation (0.57) and the ratio of prediction to deviation (1.52) for curd firmness. Similar fitting statistics of the prediction models were obtained for rennet coagulation time and curd firming time. In conclusion, the prediction models for all three coagulation traits were below the threshold to consider the prediction models adequate even for rough screening of the samples.

Rapid and High-Performance Analysis of Total Nitrogen in Coco-Peat Substrate by Coupling Laser-Induced Breakdown Spectroscopy with Multi-Chemometrics

Nitrogen is an important nutrient element for crop growth. Rapid and accurate acquisition of nitrogen content in cultivation substrate is the key to precise fertilization. In this study, laser-induced breakdown spectroscopy (LIBS) was used to detect the total nitrogen (TN) of coco-peat substrate. A LIBS spectrum acquisition system was established to collect the spectral line signal of samples with wavelengths ranging from 200 nm to 860 nm. Synergy interval partial least squares (Si-PLS) algorithm and elimination of uninformative variables (UVE) algorithm were used to select the spectral data of TN characteristic lines in coco-peat substrate. Univariate calibration curve and partial least squares regression (PLSR) were used to build mathematical models for the relationship between the spectral data of univariate characteristic spectral lines, full variables and screened multi-variable characteristic spectral lines of samples and reference measurement values of TN. By comparing the detection performance of calibration curves and multivariate spectral prediction models, it was concluded that UVE was used to simplify the number of spectral input variables for the model and PLSR was applied to construct the simplest multivariate model for the measurement of TN in the substrate samples. The model provided the best measurement performance, with the calibration set determination coefficient () and calibration set root mean square error (RMSEC) values of 0.9944 and 0.0382%, respectively; the prediction set determination coefficient () and prediction set root mean square error (RMSEP) had values of 0.9902 and 0.0513%, respectively. These results indicated that the combination of UVE and PLSR could make full use of the variable information related to TN detection in the LIBS spectrum and realize the rapid and high-performance measurement of TN in coco-peat substrate. It would provide a reference for the rapid and quantitative assessment of nutrient elements in other substrate and soil.

Detection of dried jujube from fresh jujube with different variety and maturity after hot air drying based on hyperspectral imaging technology

The consistent quality of dried jujube has a significant impact on its economic value and consumer acceptance. Currently, the evaluation of dried jujube quality heavily relies on agricultural experts and advanced instrument analysis, which is time costly and is limited by many uncertainties. This study aims to develop a rapid and accurate method for classifying dried jujube produced from fresh jujube at different stages of maturity using hyperspectral imaging. Dried jujube samples were collected from two varieties (Zhongyang Muzao jujube and Yuanling jujube) in Hebei Province, representing two maturity stages: Kimri and Rutab. Linear partial least squares-discriminant analysis (PLS-DA), K-Nearest Neighbor (KNN) and support vector machine (SVM)algorithms were combined with six pre-processing methods. the pretreatment spectra of Area Naturalization (AN), Baseline and Multiple scattering correction (MSC) were selected for further analysis. Successive projections algorithm (SPA), Competitive adaptive reweighted sampling （CARS）and Uninformative Variables Elimination (UVE)were combined to select characteristic variables. A total of 16 characteristic variables were selected by AN-UVE-SPA-SVM model. This model achieved the highest accuracies of 96.0 % and 93.1 % in the final discriminant accuracy for the calibration and Validation sets, respectively. The hardness, soluble solid content (SSC), and moisture of two categories of dried jujube derived from fresh jujube at various stages of maturity were analyzed. The results showed that dried jujube produced from fresh jujubes at the Rutab stage were more likely to exhibit characteristics of high SSC and low hardness' compared to those produced at the Kimri stage. The partial least squares regression (PLSR) model established based on characteristic wavelengths extracted by AN-UVE-SPA showed the best results in predicting SSC and moisture. This study showcased the feasibility of identifying and classifying dried jujube using hyperspectral imaging technology, laying a solid foundation for further research in the rapid non-destructive testing of dried fruits and vegetables through the application of hyperspectral technology.

Electricity Price Forecasting in the Irish Balancing Market

Short-term electricity markets are becoming more relevant due to less-predictable renewable energy sources, attracting considerable attention from the industry. The balancing market is the closest to real-time and the most volatile among them. Its price forecasting literature is limited, inconsistent and outdated, with few deep learning attempts and no public dataset. This work applies to the Irish balancing market a variety of price prediction techniques proven successful in the widely studied day-ahead market. We compare statistical, machine learning, and deep learning models using a framework that investigates the impact of different training sizes. The framework defines hyperparameters and calibration settings; the dataset and models are made public to ensure reproducibility and to be used as benchmarks for future works. An extensive numerical study shows that well-performing models in the day-ahead market do not perform well in the balancing one, highlighting that these markets are fundamentally different constructs. The best model is LEAR, a statistical approach based on LASSO, achieving a mean absolute error of 32.82 €/MWh, surpassing more complex and computationally demanding approaches with errors ranging from 33.71 €/MWh to 44.55 €/MWh.

Nondestructive detection of Clonorchis sinensis infection of raw Pseudorasbora parva fish by near-infrared hyperspectral imaging

Fish parasites adversely affect aquaculture and can be transmitted to humans via the food chain, necessitatingrapid, non-destructive detection methods. This research investigated the use of visible/near-infrared (VIS/NIR) hyperspectral imaging for the nondestructive identification of Clonorchis sinensis infected Pseudorasbora parva fish, with a 31.82% infection rate in the wild in China. Our analysis of various preprocessing, feature extraction, and modeling algorithm combinations revealed that using standard normal variate preprocessing, competitive adaptive reweighted sampling for feature selection, and partial least squares discriminant analysis significantly enhanced detection efficacy. Detection accuracies reached 99.24% and 93.93% for calibration and validation sets, respectively, based on the spectrum of whole fish. Detection accuracy varied among different fish parts, with the highest in the stomach, followed by the head, caudal fin, and back. Specifically, Stomach spectra achieved 99.24% and 93.93% accuracy for calibration and validation sets, respectively. Notably, a single local spectrum from a 15 × 15 pixel-sized area of the stomach detected parasites with 99.99% and 90.90% accuracy for calibration and validation sets, respectively, significantly increasing detection speed. This study provides a comprehensive methodology for the rapid, nondestructive detection of fish parasites via VIS/NIR hyperspectral imaging technology, demonstrating significant potential for application.

Spectrophotometric determination of celecoxib and tramadol in the new approved formulated dosage form using principle component regression assistive model

Celecoxib and tramadol have been combined in a novel FDA-approved medication to address acute pain disorders requiring opioid treatment when other analgesics proved either intolerable or ineffective. The absorbance spectra of celecoxib and tramadol exhibit significant overlap, posing challenges for their individual quantification. This study introduces a spectrophotometric quantification approach for celecoxib and tramadol using a principle component regression assistive model to assist resolving the overlapped spectra and quantifying both drugs in their binary mixture. The model was constructed by establishing calibration and validation sets for the celecoxib and tramadol mixture, employing a five-level, two-factor experimental design, resulting in 25 samples. Spectral data from these mixtures were measured and preprocessed to eliminate noise in the 200–210 nm range and zero absorbance values in the 290–400 nm range. Consequently, the dataset was streamlined to 81 variables. The predicted concentrations were compared with the known concentrations of celecoxib and tramadol, and the errors in the predictions were evidenced calculating root mean square error of cross-validation and root mean square error of prediction. Validation results demonstrate the efficacy of the models in predicting outcomes; recovery rates approaching 100 % are demonstrated with relative root mean square error of prediction (RRMSEP) values of 0.052 and 0.164 for tramadol and celecoxib, respectively. The selectivity was further evaluated by quantifying celecoxib and tramadol in the presence of potentially interfering drugs. The model demonstrated success in quantifying celecoxib and tramadol in laboratory-prepared tablets, producing metrics consistent with those reported in previously established spectrophotometric methods.

Quantitative Structure–Activity Relationship Models for the Angiotensin-Converting Enzyme Inhibitory Activities of Short-Chain Peptides of Goat Milk Using Quasi-SMILES

The inhibitory activity of peptides on angiotensin-converting enzyme (ACE) is a measure of their antihypertensive potential. Quantitative structure–activity relationship (QSAR) models obtained based on the analysis of sequences of amino acids are suggested. The average determination coefficient for the active training sets is 0.36 ± 0.07. The average determination coefficient for validation sets is 0.79 ± 0.02. The paradoxical situation is caused by applying the vector of ideality of correlation, which improves the statistical quality of a model for the calibration and validation sets but is detrimental to the statistical quality of models for the training sets.

Protein content prediction of rice grains based on hyperspectral imaging

This study utilized hyperspectral imaging technology combined with mathematical modeling methods to predict the protein content of rice grains. Firstly, the Kjeldahl method was used to determine the protein content of rice grains, and different preprocessing techniques were applied to the spectral information. Then, a prediction model for rice grain protein content was developed by combining the spectral data with the protein content. After performing multiplicative scatter correction (MSC) preprocessing and selecting feature wavelengths based on successive projections algorithm (SPA), the multivariate linear regression (MLR) model showed the best prediction performance, with a calibration set R2C of 0.9393, a validation set R2V of 0.8998, an RMSEV of 0.1725, and an RPD of 3.16. Finally, the quantitative protein content model was mapped pixel by pixel to visualize the distribution of rice protein, providing possibilities for non-destructive protein content detection.

Calibration and context in human evaluation of machine translation

Abstract Human evaluation of machine translation is considered the “gold standard” for evaluation, but it remains a challenging task for which to define best practices. Recent work has focused on incorporating intersentential context into human evaluation, to better distinguish between high-performing machine translation systems and human translations. In this work, we examine several ways that such context influences evaluation and evaluation protocols. We take a close look at annotator variation through the lens of calibration sets and focus on the implications for context-aware evaluation protocols. We then demonstrate one way in which degraded target-side intersentential context can influence annotator scores of individual sentences, a finding that supports the context-aware approach to evaluation and which also has implications for best practices in evaluation protocols.

Prediction Accuracy of Soil Chemical Parameters by Field- and Laboratory-Obtained vis-NIR Spectra after External Parameter Orthogonalization.

One challenge in predicting soil parameters using in situ visible and near infrared spectroscopy is the distortion of the spectra due to soil moisture. External parameter orthogonalization (EPO) is a mathematical method to remove unwanted variability from spectra. We created two different EPO correction matrices based on the difference between spectra collected in situ and, respectively, spectra collected from the same soil samples after drying and sieving and after drying, sieving and finely grinding. Spectra from 134 soil samples recorded with two different spectrometers were split into calibration and validation sets and the two EPO corrections were applied. Clay, organic carbon and total nitrogen content were predicted by partial least squares regression for uncorrected and EPO-corrected spectra using models based on the same type of spectra ("within domain") as well as using laboratory-based models to predict in situ collected spectra ("cross-domain"). Our results show that the within-domain prediction of clay is improved with EPO corrections only for the research grade spectrometer, with no improvement for the other parameters. For the cross-domain predictions, there was a positive effect from both EPO corrections on all parameters. Overall, we also found that in situ collected spectra provided an equally successful prediction as laboratory-based spectra.

The assessment of soil organic matter in the KwaZulu-natal province of South Africa and its relationship to spectroscopy data

Soil organic matter (SOM) is made up of decomposing biotic material in various stages, as well as compounds generated by plant roots and soil organisms—it helps the soil’s biophysical functions. Laboratory spectroscopy, for example, provides a novel technique to analyse SOM content because it is both cost and time-efficient. Because of its composition and biophysical features, SOM has a distinct spectral reflectance; this relationship has been effectively exploited to estimate and predict SOM. The purpose of this study is to investigate the link between SOM concentration and laboratory-based soil spectral reflectance in the Emakhosaneni subplace of uThukela District, KwaZulu-Natal, South Africa. We collected 13 random soil samples from each of the four major land use types (agricultural, residential, eroded land, and rangeland), totalling 52 samples. To assess the SOM content, we oven-dried the materials overnight (105 °C), crushed them, sieved them (1 mm), and analysed them using the Walkley-Black method. The spectral reflectance of the soil samples was then acquired using the Analytical Spectral Device (ASD) FieldSpec3 and pre-processed for noise reduction before analysis. We found that the area’s average SOM content was 2.19%, with agricultural land having the greatest average SOM content (2.98%), followed by rangeland (2.46%), residential (1.77%), and eroded land area (1.56%). However, the average reflectance of the spectra was higher on eroded areas and decreased to a minimum on agricultural areas. This is attributed to the relationship between soil colour and SOM. Correlation analysis demonstrated a moderately negative relationship between SOM concentration and spectral reflectance over the whole spectral range covered (400–2400 nm). Our partial least squares regression analysis revealed that pre-processed spectra data models (FDT; R2 = 0.46 and SG; R2 = 0.34) performed better than raw data models, in both calibration and validation data sets, respectively. Despite the influence of noise on raw data model performance, they modestly predicted SOM content, particularly on the validation data set. However, the developed model cannot be used due to the very low coefficient of determination. Our results highlight the importance of spectroscopy to assess SOM content, and further research can be carried out on a larger scale.