Interpretability of selected variables and performance comparison of variable selection methods in a polyethylene and polypropylene NIR classification task.

Near infrared (NIR) spectra are collected as a high amount of absorption values which usually greatly exceeds the sample size. Variable selection methods are employed in NIR spectroscopy to avoid "curse of dimensionality" related issues. In this paper, we examined the interpretability of selected variables, that is, how much selected wavelengths are related to the chemical structure of the materials studied, and if the relation is important for classification performance. Additionally, we examined classification performance in dependence on the number of selected variables. For this purpose, relative standard deviation (RSD), successive projection algorithm (SPA), stepwise decorrelation of variables (SELECT), genetic algorithm (GA), principal component analysis (PCA), and random (RANDOM) variable selection were applied in two-class classification modelling using linear discriminant analysis (LDA) or a support vector machine (SVM). Different pre-treatments and sample sizes were considered. Variable selection improved classification performance and variables selected by a majority of the methods considered were conveniently related to chemical structure. Interpretability and performance increase/decrease depend greatly on the number of selected variables, however. Since selected variables reveal great chemical interpretability, some variable selection methods could be employed to determine material characteristic absorption bands. SELECT and SPA displayed the best properties among the methods considered. To avoid faulty results, optimization of the number of selected variables should become the crucial stage in the variable selection process.

The Distribution of Minerals in Crucial Plant Parts of Various Elderberry (Sambucus spp.) Interspecific Hybrids.

In view of growing requirements of the food industry regarding elderberries (genus Sambucus), a need to increase their productivity and improve their chemical composition has emerged. With this purpose in mind, numerous elderberry interspecific hybrids have been created. In the present work, the content of minerals in their crucial plant parts was studied. It was also investigated whether superior genotypes regarding the mineral composition of berries and inflorescences could be predicted at early stages of plant development. The results showed that elderberry leaves contained the highest amounts of Ca, Mg, Mn, Zn, and Sr, while K and P were predominant in fruit stalks. Fe and Al prevailed in roots and Cu in bark. Although berries showed lower mineral content compared to other plant parts, their mineral content is not negligible and could be comparable to other commonly consumed berries. Genotypes with a favorable mineral content of inflorescences and berries could be predicted on the basis of known mineral composition of their shoots and leaves. The study also indicates that S. nigra genotypes and the majority of interspecific hybrids analyzed are suitable for further genetic breeding or cultivation.

Effects of integrated and organic pest management with copper and copper-free preparations on tomato (Lycopersicum esculentum Mill.) fruit yield, disease incidence and quality.

The present study investigates the effect of exclusion of copper preparations from spray programs for disease control in integrated and organic tomato production systems on disease incidence, yield and fruit vitamin C, α-tocopherol and carotenoids content for Monroe, Optima and Paki cultivars. Open field trials were conducted in a randomized block system with four repetitions over two years. Treatment for plant disease control comprised four spraying programs, two within each production system. Yields were significantly affected by cultivar, pest management system and their interaction. Copper-free integrated and organic spraying programs were almost as effective in disease control as standard treatments which included copper preparations. The spraying program significantly affected the contents of vitamin C, α-tocopherol, phytoene, and ß-carotene in fruits; however, the effect was cultivar related. The lycopene contents appeared to be unaffected by the spraying program or growing system.