The integration of Large Language Models (LLMs) with web scraping and crawling techniques is transforming automated web data extraction by enabling semantic understanding and adaptability. This Systematic Literature Review (SLR) synthesizes evidence regarding this integration, focusing on tools, models, challenges, evaluation methods, trends, and applications. Following PRISMA guidelines, we conducted a rigorous search across Scopus, Web of Science, ACM, and IEEE databases (2021–2025). From 976 screened records, 91 high-quality studies (53 conference papers and 38 journal articles) were selected after duplicate removal, screening, and AI-powered quality assessment. The field has experienced explosive growth, with 84
This study examines the effects of a three-month pedagogical intervention that integrated artificial intelligence (AI), social media, and web-based tools to strengthen digital literacy, creativity, and cultural participation among secondary education students in Ecuador. The intervention was theoretically grounded in perspectives of inclusive digital education and Universal Design for Learning (UDL), emphasizing participation, accessibility, and collaborative knowledge construction. The intervention involved 61 students supported by 31 university facilitators and was developed under a mixed-methods action research design with a pre–post (quasi-experimental) approach. Pre- and post-test surveys were administered to assess changes in digital competencies and creativity, while semi-structured interviews explored students’ perceptions of creative expression and their engagement with the cultural and technological ecosystem. Quantitative results showed statistically significant improvements in digital literacy and creativity (p < 0.001), while qualitative findings evidenced increased student empowerment, critical awareness of algorithms, and active cultural participation. The integration of AI and social media promoted an inclusive, student-centered learning environment that enhanced autonomy, reflective thinking, and media engagement. These results suggest that hybrid and culturally contextualized AI-mediated interventions may foster 21st-century competencies, strengthen digital equity, and promote creative agency in educational contexts of the Global South, particularly within emerging digital learning environments in Ecuador.
The global citrus-processing industry generates 15–32 million tonnes of waste annually. Lemon-processing residues—peels, seeds, and pomace—constitute 45–55% of fruit mass and harbour high-value bioactive compounds amenable to cascade valorisation. This review (Part I of a two-part series) examines green extraction technologies for recovering bioactive compounds from lemon waste streams. Following bibliometric analysis of 847 publications (2003–2025), this work delineates the compositional heterogeneity of lemon fractions and establishes a hierarchical framework for value-added products encompassing essential oils, pectin, polyphenols, seed oils, citric acid, industrial enzymes, α-cellulose, and nanocrystalline cellulose. Four sustainable extraction methodologies are systematically evaluated: ultrasound-assisted extraction, microwave-assisted extraction, supercritical CO2 extraction, and enzyme-assisted extraction. Comparative assessment demonstrates yield improvements of 16–112% over conventional approaches, processing-time reductions of 89–98%, and energy savings up to 95%. Critical research gaps include fragmented single-product valorisation, insufficient techno-economic assessment, and limited industrial-scale validation. Integrated cascade biorefineries employing sequential green extraction protocols offer economically viable pathways for transforming lemon waste into diversified revenue streams. Industrial implementation, circular-economy integration, and techno-economic feasibility are addressed in Part II.
Lactiplantibacillus plantarum strains are increasingly recognized for their combined probiotic and antimicrobial activities, offering potential applications in gut health management and pathogen control. This study characterized the intracellular (Met-Int) and extracellular (Met-Ext) metabolomic profiles of L. plantarum UTNGt2 (Gt2), UTNGt3 (Gt3), and UTNGt28L (Gt28L) isolated from tropical fruits and evaluated their probiotic, antimicrobial, cytotoxic, and immunomodulatory properties in vitro. Metabolomic profiling was performed using liquid chromatography–tandem mass spectrometry (LC–MS/MS) with a SWATH (Sequential Windowed Acquisition of All Theoretical Fragment Ion Mass Spectra) acquisition method. Cytotoxicity and cell viability were assessed by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) and LDH (lactate dehydrogenase) release assays in colon epithelial cells, while cytokine responses (IL-10, IL-1β) were quantified to determine immunomodulatory effects. Antimicrobial mechanisms were examined by scanning and transmission electron microscopy (SEM/TEM) on Staphylococcus aureus ATCC1026. LC–MS/MS identified 117 Met-Int and 32 Met-Ext across the three strains, revealing shared metabolites (e.g., l-tryptophan, adenosine) and distinct strain-specific compounds (e.g., harmine, lincomycin, baicalein) associated with bioactivity. Pathway enrichment analysis indicated four enriched pathways in Gt2, eight in Gt3, and ten in Gt28L, reflecting differential specialization in amino-acid, carbohydrate, and cofactor metabolism. Gt3 exhibited the most diverse antimicrobial metabolite repertoire, whereas Gt28L showed the strongest anti-inflammatory effect, increasing IL-10 secretion by 6.5-fold and reducing IL-1β by 50
Weed identification and quantification are processes that are usually manual, subjective, and error-prone. Weeds compete with crops for nutrients, minerals, physical space, sunlight, and water. Thus, weed identification is a crucial component of precision agriculture for autonomous removal and site-specific treatments, efficient weed control, and sustainability. Convolutional Neural Networks (CNNs) are very common in weed identification. This work implemented CNN models for semantic segmentation based on the U-Net architecture for automatically segmenting and quantifying weeds in potato crops using RGB images acquired by a drone at 9–10 m height, flying at 1 m/s. Remote sensing images are affected by factors that degrade image quality and the model’s accuracy. Five U-Net variants were evaluated: the original U-Net, Residual U-Net, Double U-Net, Modified U-Net, and AU-Net. The models were trained using the TensorFlow/Keras frameworks on Google Colab Pro+, following the Knowledge Discovery in Databases (KDD) methodology for image analysis. Each model was trained using a diverse custom dataset in uncontrolled environments, considering six classes: background, Broadleaf dock (Rumex obtusifolius), Dandelion (Taraxacum officinale), Kikuyu grass (Cenchrus clandestinum), other weed species, and the crop potato (Solanum tuberosum L.). The models’ segmentation was widely assessed using Mean Dice Coefficient, Mean IoU, and Dice Loss metrics. The results showed that the Residual U-Net model performed the best in multi-class segmentation, achieving a Mean IoU of 0.8021, a performance comparable to or superior to that reported by other authors. Additionally, a Student’s t-test was applied to complement the data analysis, suggesting that the model is reliable for weed quantification.
