Conference Agenda

Algorithmic design and theories of spatial organization in video games find a common ground in solving one of the problems in the pre-feasibility stage of geometric road design: connecting two points on a terrain through the most optimal and efficient route. This preliminary solution is a conjunction of multiple fields of knowledge, which come together to develop a tool that materializes these theories in a practical case of study. From this development, a first algorithmic design prototype emerges, presenting the Preliminary Designs of the Zero-Line Method across nine routes that are analyzed and compared to identify the characteristics of those that approach the destination point through algorithmic design. Future work aims to refine the approximation of these Preliminary Designs in order to proceed with the subsequent pre-feasibility stages that will complete the solution to the optimization problem in the manual procedures of the zero-line method.

Monitoring bat echolocation using ultrasound microphones is essential in ecological studies due to its non-invasive nature and wide spatial-temporal coverage. However, the large volume of data and high temporal resolution of the signals pose challenges in terms of management, storage, visualization, and annotation for scientific analysis. These tasks are limited by the dispersion of information across multiple storage media and dependence on proprietary software, which hinders collaborative work and increases research costs. This work presents the design and development of a web platform for high-resolution spectral visualization and collaborative taxonomic annotation of ultrasonic echolocation signals. A software engineering approach was adopted, implementing a decoupled architecture using Django REST Framework in the backend for robust data management and Next.js in the frontend to ensure a responsive user interface, integrating the Web Audio API and rendering with HTML5 Canvas, allowing high-resolution digital signal processing directly in the browser without loss of quality. As a result, a functional prototype was developed for the online visualization of echolocation signal spectrograms, enabling the recording of taxonomic and behavioral information, automatic signal processing and the extraction of spectral features, facilitating collaborative work in bat bioacoustics.

Th Literature review is a core component of the scientific research process; however, the rapid growth of academic production has exceeded the capacity of traditional manual approaches. In response to this challenge, computational tools oriented toward automation and reproducibility have gained increasing relevance in bibliometric analysis. In this context, this paper introduces redalycR, an R package designed to automate bibliometric analysis through a reproducible architecture based exclusively on open data from the Redalyc database.
The paper describes the conceptual design and implementation of the package, as well as its integration into analytical workflows in R for the acquisition, cleaning, normalization, and structuring of bibliographic metadata. Through an illustrative application case, it is shown how redalycR enables the programmatic and reproducible generation of descriptive keyword analyses and keyword co-occurrence networks. The results demonstrate that the package facilitates the structural exploration of Spanish-language scientific corpora, complementing existing bibliometric tools and expanding analytical capabilities over open-access literature. Finally, the current limitations of the package are discussed, and a future development agenda aimed at its extension and strengthening is outlined.

Financial statement fraud poses a threat to the transparency of financial information reported in the markets and to investor confidence. Given this scenario, the development and implementation of AI-based systems that incorporate ML techniques allow for the analysis of large volumes of financial data and the early detection of anomalies and patterns associated with fraudulent practices in financial reports. The CRISP-ML(Q) methodology is used, along with a database of 146,045 records from the SEC and COMPUSTAT, with a marked imbalance of fraud cases (0.00066%). Supervised (random forest, RF; logistic regression, LR) and unsupervised (isolation forest, IF; one-class SVM; local outlier factor, LOOF) models were used, along with feature selection, normalization, and oversampling with SMOTE. The results show that the RF model achieved a high overall accuracy of 98%, standing out for its ability to correctly classify most observations. Unsupervised models achieved a high level of recall in fraud detection (80%), demonstrating their effectiveness in identifying a greater number of fraudulent cases. It is concluded that effective fraud detection requires more robust approaches that combine traditional techniques with cost-sensitive algorithms, adaptive thresholds, and supervised models, strengthening anomaly identification and supporting financial risk auditing and management.

Inventory management in industrial settings is shaped by high variability in consumption and replenishment lead times, which hinders planning. Given the dispersion of predictive approaches reported in the literature, a systematic review helps consolidate comparable evidence. Accordingly, this study conducts a systematic literature review (SLR) aimed at characterizing recurrent models and variables used for forecasting critical supplies, with a focus on the dairy industry and AI. The review followed Fink’s methodology; from 780 initial records, 39 eligible studies were selected. Results show that the most recurrent variables are concentrated in the temporal and operational components of demand, highlighting historical consumption and seasonality, complemented by lead time and, when traceability exists, available stock. In terms of approaches, neural networks/deep learning predominate especially recurrent architectures (LSTM/GRU/RNN) while traditional machine learning methods are used as comparative baselines. For evaluation, error metrics (MAE, RMSE, MSE, MAPE) prevail; however, for intermittent consumption a dual framework is proposed that separates occurrence (Macro-F1) and magnitude (MAE). These findings guide the design and evaluation of predictive models applicable to real-world scenarios with limited data availability

Precipitation prediction plays a critical role in water resource management, agriculture, and climate risk mitigation, particularly in regions characterized by strong climatic variability, such as Ecuador. This study investigates the application of machine learning techniques to precipitation prediction using a long-term climatic dataset spanning 1950 to 2023. Three regression models were evaluated: Decision Tree Regressor, Random Forest Regressor, and Gradient Boosting Regressor. The dataset was preprocessed through temporal decomposition, logarithmic transformation of precipitation, and cyclical encoding of seasonal effects to capture long-term trends and annual variability. Model performance was assessed using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the coefficient of determination (R²). Results show the Gradient Boosting Regressor achieving the best performance (MAE = 27.63 mm, RMSE = 37.29 mm, R² = 0.84). Scenario-based predictions and sensitivity analysis further demonstrate the model’s physical consistency and practical applicability. The findings confirm the potential of machine learning models, particularly gradient boosting, as reliable tools for precipitation prediction and exploratory climate analysis in Ecuador.

The oil and gas sector is responsible for producing 25% of methane generated by human activities worldwide, contributing to global warming at a rate 80 times greater than CO₂. In the Ecuadorian Amazon, oil exploitation is a source of greenhouse gases of approximately 450 million cubic feet per day. This study aims to conduct a literature review on petroleum gas flaring in eastern Ecuador by systematizing documents obtained from the Google Scholar database to generate sustainable strategies based on the Ecuadorian regulatory framework. To this end, a three-phase methodological design was adopted: i) Collection of bibliographic information, ii) Systematic literature review, and iii) Design of mitigation strategies. The results show direct impacts on air quality and nearby communities, associated with more than 400 active flares and 295 unreported flaring points. Furthermore, deficiencies in outdated infrastructure and limited adoption of clean technologies are evident. Therefore, the proposed strategies integrate social and health measures, gas liquids recovery with an estimated 30% increase in LPG production, and energy recovery with a potential of 70 MWh. The findings demonstrate that implementing a comprehensive mitigation plan, aligned with national regulations and international best practices, is technically and environmentally feasible, constituting a concrete alternative for reducing atmospheric impacts and moving towards more efficient and responsible energy management in the Ecuadorian Amazon.