Objective. The scientific performance of Peruvian universities was examined and characterized by aggregating information from their institutional repositories. Design/Methodology/Approach. A protocol was followed that included the following stages: (1) analysis and optimization of the DSpace database model; (2) data recovery and validation with the OAI-PMH Protocol; (3) obtaining an index of resources available in the repositories; (4) metadata extraction and processing; and (5) extraction and inclusion of keywords in the database. Results/Discussion. A diverse use of keywords was identified. 69,1% of them appeared only once. A significant gap was evident in producing theses between public and private universities. Conclusions. The discrepancy between the research areas of Peruvian universities and regional needs highlights the importance of greater alignment between academic production and the demands of society in each region; this underlines the need to adapt research programs to address specific problems and contribute more effectively to regional development.
26 September 2023 in ICST Transactions on Scalable Information Systems
INTRODUCTION: The emerging industrialization of software promotes the continuous development of new evaluation models that adapt to the product's characteristics and the organizations' needs. OBJECTIVES: This article aims to identify the software quality models proposed between 2016 and 2020. METHODS: We conducted an exploratory systematic review in the SciELO, IEEE Xplore, and Scopus databases, resulting in 14 research papers proposing new models for assessing software quality. RESULTS: 79% of the research papers were extracted from IEEE Xplore, and 50% of them were authored by individuals affiliated with European institutions. We presented various software quality models focused on assessing both product quality and usability for specific purposes. Additionally, we reported that the proposed models are based on common software quality metrics standards such as CMMI, ISO/IEC 9126, and others. CONCLUSION: The review serves to update the state of the art regarding the software quality models proposed in recent years, which will be valuable for researchers and software developers when seeking evaluation models that align with their specific needs.
20 July 2022 in Revista Científica de Sistemas e Informática
Poseer competencias digitales es una necesidad que las Mypes están obligadas a tener en el mercado actual cambiante, en donde el comercio se ha visto influenciado por las TIC. En este sentido, el objetivo del estudio fue determinar el nivel de las competencias digitales de las Mypes peruanas de la región San Martín en época de pos covid-19. Para ello, se realizó una investigación básica de nivel descriptivo y diseño no experimental transversal. Participaron 63 Mypes, las cuales respondieron una encuesta validada y confiabilizada. Los resultados mostraron que el nivel de conocimiento del funcionamiento genérico de las herramientas digitales fue medio en 51%, alto en 30 % y bajo en 19 %; el nivel de conocimiento de las herramientas digitales fue medio en 38%, alto en 33% y bajo en 29 %; el nivel de uso de las herramientas digitales fue medio en 63%, alto en 24% y bajo en 13%; además el nivel de acceso a la información digital fue medio en 48%, bajo en 35% y alto en 17%. Concluyendo que el nivel de competencias digitales en Mypes de San Martín es regular.
In the digital age, preserving and effectively retrieving historical academic records has a significant challenge, especially when these documents only exist in deteriorated physical formats. We propose an approach to recover data from scanned documents of grade records, by using image processing and Natural Language Processing (NLP) to enhance the accuracy of Optical Character Recognition (OCR) in these documents, essential for the preservation of digital records. Our three-step methodology: first, improves the quality of the scanned image; then, extracts text using OCR and NLP techniques to retrieve data from old physical grade cards; and finally, the extracted data is corrected using Chat-GPT and prepared for upload. The results are assuring, showing an impressive Character Error Rate (CER) of 2.15% and a Word Error Rate (WER) of 7.05%, demonstrating the high accuracy of the OCR system used and its ability to precisely extract text from scanned documents. These low error rates achieved, as a result to the successful implementation of pre-processing and post-processing techniques, as well as the use of an advanced OCR tool, underscore the potential of this OCR approach to effectively extract information from documents.
Orange waste, generally discarded, is a source of many bioactive compounds that could be used for the development of high-value-added products in the food, cosmetic, and pharmaceutical industries. The objective of this study was to evaluate the influence of extraction method (automated Soxhlet extraction and temperature-controlled maceration), solvent system, and extraction time on the recovery of bioactive compounds from Valencia orange (Citrus sinensis) by-products. Proximate characterization of the dried orange residue (DOR) was performed prior to extraction. The type of solvent (ethanol and methanol), solvent:water ratio (50, 75, and 100%), and extraction time (60 and 120 min) were evaluated in terms of total extraction yield (TEY), total phenolic content (TPC), and antioxidant capacity determined by ABTS and DPPH assays, for each extraction method. ASE generally provided higher extraction yield and total phenolic content, particularly with 75:25 ethanol:water at 120 min, whereas TCM combined with methanol produced the highest antioxidant capacity. Extracts with up to 46.32% TEY, 5.57 mg GAE/g dm, and antioxidant capacities of 66.49 and 11.10 µmol TE/g dm determined by ABTS and DPPH assays, respectively, were obtained. The results demonstrated that Valencia orange by-products are a source of phenolic compounds and antioxidants with potential for product development across different industrial sectors.
20 January 2026 in Revista Científica de Sistemas e Informática
This study analyzed the application of deep learning for automated phytosanitary detection in cacao using computer vision, comparing the performance of three architectures: ResNet50, EfficientNet-B0, and Vision Transformer (ViT-B/16). A reproducible pipeline was implemented, integrating image preprocessing, five-fold stratified cross-validation, and inferential statistical analysis using repeated-measures ANOVA. The dataset consisted of 4,390 RGB images of cacao fruits distributed across three imbalanced classes: Healthy, Black Pod Rot, and Pod Borer. All models were fully fine-tuned and trained using the AdamW optimizer, early stopping, and a dynamic learning rate scheduler. The results showed mean F1-macro values above 0.96 across all architectures, with no statistically significant differences among models (F = 0.278, p = 0.7645). Training curves exhibited stable convergence and low inter-fold variability, with no evidence of overfitting. These findings indicate that system performance primarily depends on the quality of the experimental pipeline and class imbalance handling rather than on the specific architecture employed.