Software is increasingly vital, with automated systems regulating critical functions. As development demands grow, manual code review becomes more challenging, often making testing more time-consuming than development. A promising approach to improving defect detection at the source code level is the use of artificial intelligence combined with natural language processing (NLP). Source code analysis, leveraging machine-readable instructions, is an effective method for enhancing defect detection and error prevention. This work explores source code analysis through NLP and machine learning, comparing classical and emerging error detection methods. To optimize classifier performance, metaheuristic optimizers are used, and algorithm modifications are introduced to meet the study’s specific needs. The proposed two-tier framework uses a convolutional neural network (CNN) in the first layer to handle large feature spaces, with AdaBoost and XGBoost classifiers in the second layer to improve error identification. Additional experiments using term frequency–inverse document frequency (TF-IDF) encoding in the second layer demonstrate the framework’s versatility. Across five experiments with public datasets, the accuracy of the CNN was 0.768799. The second layer, using AdaBoost and XGBoost, further improved these results to 0.772166 and 0.771044, respectively. Applying NLP techniques yielded exceptional accuracies of 0.979781 and 0.983893 from the AdaBoost and XGBoost optimizers.

Software is increasingly vital, with automated systems regulating critical functions. As development demands grow, manual code review becomes more challenging, often making testing more time-consuming than development. A promising approach to improving defect detection at the source code level is the use of artificial intelligence combined with natural language processing (NLP). Source code analysis, leveraging machine-readable instructions, is an effective method for enhancing defect detection and error prevention. This work explores source code analysis through NLP and machine learning, comparing classical and emerging error detection methods. To optimize classifier performance, metaheuristic optimizers are used, and a modified algorithm is introduced to meet the study’s specific needs. The proposed two-tier framework uses a convolutional neural network (CNN) in the first layer to handle large feature spaces, with AdaBoost and XGBoost classifiers in the second layer to improve error identification. Additional experiments using Term Frequency-Inverse Document Frequency (TF-IDF) encoding in the second layer demonstrate the framework’s versatility. Across five experiments with public datasets, the CNN achieved an accuracy of 0.768799. The second layer, using AdaBoost and XGBoost, further improved these results to 0.772166 and 0.771044, respectively. Applying NLP techniques yielded exceptional accuracies of 0.979781 and 0.983893 from the AdaBoost and XGBoost optimizers.

The study area is the city of Zintan, in northwestern Libya, which has grown over the past 30 years. Its current population is roughly 30,000. Although the city is in part commercial, most of the population engages in agriculture and primarily grows cereal crops (wheat and barley). The demand of the growing city for agricultural products has increased, intensifying the use of fertilizers, pesticides, and insecticides. Consequently, concentrations of potentially toxic elements (PTE) in the soil have increased. To assess the soil quality (to provide a snapshot of the condition of the soil), systematic sampling was undertaken across a grid of about 2 km × 2 km. The main objective was to determine the chemical and mineral composition of the area of interest, keeping in mind the geological footprint of the terrain. The geologic framework itself is not conducive to elevated concentrations of elements like U, Th, Mo, As, Hg, Pb, and Cr. Therefore, metal concentrations greater than the amounts in the Earth’s crust are most likely of anthropogenic origin. A total of 143 samples were collected and chemical analyses were performed using a Thermo Fisher Scientific Niton XL3t GOLDD+ XRF analyzer for the following elements: Mo, Zr, Sr, U, Rb, Th, Pb, Au, Se, As, Hg, Zn, W, Cu, Ni, Co, Fe, Mn, Cr, V, Ti, Sc, Ca, K, S, Ba, Cs, Te, Sb, Sn, Cd, Pd, and Ag. This paper provides examples of elevated concentrations, potentially harmful to the environment, such as those of the following: sulfur of unknown origin (two to three times higher than the Earth’s crust average); arsenic, given that there are no related natural phenomena (all the samples measured displayed concentrations higher than those found in the Earth’s crust); mercury (concentrations much higher than permissible levels); cesium (additional investigations required to determine the origin); molybdenum; and uranium likely resulting from the use of superphosphates (concentrations nearly always significantly higher than those in the Earth’s crust).