CRIS
Permanent URI for this communityhttps://cris.ute.edu.ec/handle/123456789/1
Browse
3 results
Search Results
Now showing 1 - 3 of 3
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Automatic Classification of Electrical Complaints Using Decision Trees and Random Forest: A Case Study Applied to CNEL EP(Salesian Polytechnic University of Ecuador, 2026-01-01); Gloria Margarita Garcés BeltránThis study presents an automated classification system to prioritize electrical service complaints at CNEL EP. A total of 143,113 real records were processed through data cleaning, missing-value imputation, and the engineering of predictive variables reflecting complaint urgency and recurrence. Based on these criteria, the target variable “Priority” was defined to distinguish high-priority from normal complaints. Supervised learning models, specifically Decision Tree and Random Forest, were then trained using one-hot encoding and cross-validation. Random Forest delivered the best performance, achieving 91% accuracy and an AUC-ROC of 0.89. These results indicate that the proposed system can significantly improve technical resource allocation and reduce response times for electrical complaints. Moreover, the study demonstrates the feasibility of integrating machine-learning techniques into the operational management of electric distribution companies, enabling future enhancements and real-time deployment. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Random Forest modeling of bipolar affective disorder in Ecuador(AG Editor (Argentina), 2025-07-31); Zilma Diago AlfesBipolar affective disorder is a mental disorder characterized by depressive and manic or hypomanic episodes. The complexity of the diagnosis of bipolar affective disorder due to the overlapping of its symptoms with other mood disorders led researchers and doctors to search for new and advanced techniques for the precise detection of bipolar affection disorder. One of these methods is the use of advanced machine learning algorithms under a statistical methodology for building logistical regression models, Random Forest. Support vector machines, Decision Tree, K-Nearest Neighbors, and Gradient Boosting, with 146 data collected from the psychiatric services affiliated with the mental health system of Ecuador. At the inferential level, the results suggest that the implementation of automatic algorithms based on the different methodologies for building models enables the successful prediction or classification of individuals with bipolar affective disorders in Ecuador compared to controlled patients who do not profile under this pathological picture. It is the best Random Forest statistical model (89.35 %) that dictates the best performance metrics compared to the Gradient Boosting model. The evolution of the overall prevalence of bipolar affective disorders in Ecuador over the past 22 years has increased by a small differential. However, from 2020 to 2022, there has been a considerable increase in the percentage prevalence of cases of bipolar affective disorders in Ecuador. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Agent-Based Model and Machine Learning for the Analysis of Barter Regulated by Lotka-Volterra Equations in an Artificial Society(Springer Nature Switzerland, 2025-02-25)Juan VillacrésIn this paper, agent-based modeling and random forest are used to model the exchange of goods and services through barter. It is regulated by the Lotka-Volterra equation system–which is often used to analyze the dynamics of populations of different species. These equations, the ones that correspond to the case of mutualism between species, help to ensure that the amount of goods and services that a person can exchange does not accumulate during certain time among few individuals. This situation is repeated for the case in which empirical information is integrated into the modeling. Through such information, the coincidence of desires is modeled with the help of the random forest technique, which allows the modeling to be more closely related to a real situation.