Classification of Biodegradable Substances Using Balanced Random Trees and Boosted C5.0 Decision Trees.

Substances that do not degrade over time have proven to be harmful to the environment and are dangerous to living organisms. Being able to predict the biodegradability of substances without costly experiments is useful. Recently, the quantitative structure-activity relationship (QSAR) models have proposed effective solutions to this problem. However, the molecular descriptor datasets usually suffer from the problems of unbalanced class distribution, which adversely affects the efficiency and generalization of the derived models. Accordingly, this study aims at validating the performances of balanced random trees (RTs) and boosted C5.0 decision trees (DTs) to construct QSAR models to classify the ready biodegradation of substances and their abilities to deal with unbalanced data. The balanced RTs model algorithm builds individual trees using balanced bootstrap samples, while the boosted C5.0 DT is modeled using cost-sensitive learning. We employed the two-dimensional molecular descriptor dataset, which is publicly available through the University of California, Irvine (UCI) machine learning repository. The molecular descriptors were ranked according to their contributions to the balanced RTs classification process. The performance of the proposed models was compared with previously reported results. Based on the statistical measures, the experimental results showed that the proposed models outperform the classification results of the support vector machine (SVM), K-nearest neighbors (KNN), and discrimination analysis (DA). Classification measures were analyzed in terms of accuracy, sensitivity, specificity, precision, false positive rate, false negative rate, F1 score, receiver operating characteristic (ROC) curve, and area under the ROC curve (AUROC).

The Effect of a New Coating on the Drying Performance of Fruit and Vegetables Products: Experimental Investigation and Artificial Neural Network Modeling.

A study on mass transfer using new coating materials (namely alginic acid and polygalacturonic acid) during osmotic dehydration-and hence in a laboratory-scale convective dryer to evaluate drying performance-was carried out. Potato and apple samples were examined as model heat-sensitive products in this study. Results indicate that the coating material containing both alginic acid and polygalacturonic acid causes higher water loss of about 17% and 7.5% and lower solid gain of about 4% and 8%, respectively, compared to uncoated potato sample after a typical 90 min osmotic dehydration process. Investigation of drying performance using both coating materials showed a higher reduction in the moisture content of about 22% and 18%, respectively, compared with uncoated samples after the 3 h drying period. Comparisons between the two proposed coating materials were also carried out. Samples (potato) coated with alginic acid demonstrated better performance in terms of higher water loss (WL), lower solid gain (SG), and notable enhancement of drying performance of about 7.5%, 8%, and 8%, respectively, compared to polygalacturonic acid. Similar outcomes were observed using apple samples. Additionally, an accurate model of the drying process based on the experimental dataset was created using an artificial neural network (ANN). The obtained mean square errors (MSEs) for the predicted water loss and solid gain outputs of the potato model were 4.0948e-5 and 3.924e-6, respectively. However, these values for the same parameters were 3.164e-5 and 4.4915e-6 for the apple model. The coefficient of determination (r2) values for the two outputs of the potato model were found to be 0.99969 and 0.99895, respectively, while they were 0.99982 and 0.99913 for the apple model, which reinforces the modeling phase.

Data on fuzzy logic based-modelling and optimization of recovered lipid from microalgae.

This article presents the data of recovered lipid from microalgae using fuzzy logic based-modelling and particle swarm optimization (PSO) algorithm. The details of fuzzy model and optimization process were discussed in our work entitled "Application of Fuzzy Modelling and Particle Swarm Optimization to Enhance Lipid Extraction from Microalgae" (Nassef et al., 2019) [1]. The presented data are divided into two main parts. The first part represents the percentage of recovered lipid using fuzzy logic model and ANOVA. However, the second part shows the variation of the cost function (recovered lipid) for the 100 runs of PSO algorithm during optimization process. These data sets can be used as references to analyze the data obtained by any other optimization technique. The data sets are provided in the supplementary materials in Tables 1-2.

Dataset on fuzzy logic based-modelling and optimization of thermophysical properties of nanofluid mixture.

This article presents the dataset generated during the process of enhancing the thermophysical properties of nanofluid mixture through fuzzy logic based-modelling and particle swarm optimization (PSO) algorithm. The details of fuzzy model and optimization phases were discussed in our work entitled "Fuzzy modeling and optimization for experimental thermophysical properties of water and ethylene glycol mixture for Al2O3 and TiO2 based nanofluids" (Said et al., 2019). In (Said et al., 2019), the detail of the numerical data has not been clearly presented. However, in this article the inputs' data values for the density, viscosity, and thermal conductivity, used for training and testing of the fuzzy model, have been mentioned which is very essential if the model has to be rebuilt again. Furthermore, the resulting data variation of the cost function for the 100 runs during the optimization process that had not been presented in (Said et al., 2019) is presented in this work. These data sets can be used as references to analyze the data resulting from any other optimization technique. The datasets are provided in the supplementary materials in Tables 1-4.

Fuzzy modeling and parameters optimization for the enhancement of biodiesel production from waste frying oil over montmorillonite clay K-30.

Transesterification is a promising technology for the biodiesel production to provide an alternative fuel that considers the environmental concerns. From the economic and environmental protection points of view, utilization of waste frying oil for the production of biodiesel addresses very beneficial impacts. Production of higher yield of biodiesel is a challenging process in order to commercialize it with a lower cost. The current study focuses on the influence of different parameters such as reaction temperature (°C), reaction period (min), oil to methanol ratio and amount of catalyst (wt%) on the production of biodiesel. The main objective of this work is to develop a model via fuzzy logic approach in order to maximize the biodiesel produced from waste frying oil using montmorillonite Clay K-30 as a catalyst. The optimization for the operating parameters has been performed via particle swarm optimization (PSO) approach. During the optimization process, the decision variables were represented by four different operating parameters: temperature (40-140 °C), reaction period (60-300 min), oil/methanol ratio (1:6-1:18) and amount of catalyst (1-5 wt%). The model has been validated with the experimental data and compared with the optimal results reported based on other optimization techniques. Results showed the increment of biodiesel production by 15% using the proposed strategy compared to the earlier study. The obtained biodiesel production yield reached 93.70% with the optimal parameters for a temperature at 69.66 °C, a reaction period of 300 min, oil/methanol ratio of 1:9 and an amount of catalyst of 5 wt%.

Improving the environmental impact of palm kernel shell through maximizing its production of hydrogen and syngas using advanced artificial intelligence.

Fossil fuel depletion and the environmental concerns have been under discussion for energy production for many years and finding new and renewable energy sources became a must. Biomass is considered as a net zero CO2 energy source. Gasification of biomass for H2 and syngas production is an attractive process. The main target of this research is to improve the production of hydrogen and syngas from palm kernel shell (PKS) steam gasification through defining the optimal operating parameters' using a modern optimization algorithm. To predict the gaseous outputs, two PKS models were built using fuzzy logic based on the experimental data sets. A radial movement optimizer (RMO) was applied to determine the system's optimal operating parameters. During the optimization process, the decision variables were represented by four different operating parameters. These parameters include; temperature, particle size, CaO/biomass ratio and coal bottom ash (CBA) with their operating ranges of (650-750 °C), (0.5-1 mm), (0.5-2) and wt% (0.02-0.10), respectively. The individual and interactive effects of different combinations were investigated on the production of H2 and syngas yield. The optimized results were compared with experimental data and results obtained from Response Surface Methodology (RSM) reported in literature. The obtained optimal values of the operating parameters through RMO were found 722 °C, 0.92 mm, 1.72 and 0.06 wt% for the temperature, particle size, CaO/biomass ratio and coal bottom ash, respectively. The results showed that syngas production was significantly improved as it reached 65.44 vol% which was better than that obtained in earlier studies.