José Adalberto Soto Mejía

José Adalberto Soto Mejía

Auto-generated profile, based on publicly available information

José Adalberto Soto Mejía

Faculty of Business Sciences, Universidad Tecnológica de Pereira, Cra 27 N° 10-02 Los Álamos, Pereira 660003, Colombia

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 February 2016 in DYNA
Verified
Journal Article
Open access
PDF file
MMMarcela María Morales-Chávez
Marcela María Morales-Chávez
JSJosé A. Soto-Mejía
José A. Soto-Mejía
WSWilliam Sarache
William Sarache
Due to opportunities for economic and social development in the biofuels market, improvement to the supply chain has become a relevant matter. In agro-industrial supply chains, procurement costs are highly relevant. Since sugarcane is a high performance raw material for ethanol production, this paper proposes a Mixed-Integer Linear Programming Model for cost optimization for harvesting, loading and transportation operations. The model determines the quantity of machines and workers to meet the biofuel plant requirements. Costs of resources for harvesting and loading as well as transportation costs from the land parcel to the production plant are minimized. Also, the model calculates the cost of penalties for shortages (unmet demand) and the cost of equipment idle time. The implementation of the model in a Peruvian biofuels company, showed a cost reduction of around 11 % when compared to the current costs.

30 June 2020
Verified
Conference Paper
Open access
MGManuela del Pilar Gómez-Suta
Manuela del Pilar Gómez-Suta
JEJulian David Echeverry-Correa
Julian David Echeverry-Correa
JSJosé A. Soto-Mejía
José A. Soto-Mejía
The construction of ontologies from texts in Spanish is a challenge since this language lacks conceptual databases to validate abstract ontology structures as concepts and relations between them. The preceding generates the necessity of using manual evaluation by human experts; carrying high expenses that limit the calibration of algorithm parameters and large-scale evaluations. This document presents a proposal to evaluate abstract ontology structures through the task of semantic clustering of documents, without the expensive necessity of using manual evaluation or conceptual databases. The proposal is not only affordable but also applicable to model data and domains that lack structured knowledge resources. The experiments lead to the extraction and validation of the ontology structures from texts in Spanish regarding the domain of the Colombian armed conflict.

5 June 2026 in Information
Verified
Journal Article
Open access
PDF file
CZCarlos Mauricio Zuluaga-Ramírez
Carlos Mauricio Zuluaga-Ramírez
MGManuela del Pilar Gómez-Suta
Manuela del Pilar Gómez-Suta
JCJulio Cesar Chavarro-Porras
Julio Cesar Chavarro-Porras
SMSandra Estrada Mejia
Sandra Estrada Mejia
JSJosé Soto-Mejía
José Soto-Mejía
This article aims to propose a unifying conceptualization for the leadership domain utilizing ontology engineering to address conceptual ambiguity and information overlap. The approach involves identifying and defining structural, common, and underlying elements to organize leadership information. The methodology is based on a literature review guided by principles of ontology engineering. It focuses on collecting, analyzing, disambiguating, and synthesizing information to represent general elements in the leadership domain as categories, subcategories, and instances, including their relationships. The result is a unifying conceptualization comprising defined categories, subcategories, general instances, and relationships. This structure creates a taxonomic framework to organize the leadership’s domain information, classifying approaches, theories, models, styles, variables, characteristics, and organizational results. Expert validation confirms their suitability to represent the leadership domain in a general manner. Researchers can use it to understand and compare information, identify conceptual similarities and differences, and support knowledge dissemination within the leadership domain, reducing ambiguity and overlap.

15 November 2016 in Ingeniería y Ciencia
Verified
Journal Article
Open access
PDF file
RRRafael Renteria
Rafael Renteria
JMJosé Adalberto Soto Mejía
José Adalberto Soto Mejía
En esta investigación se propone la construcción de un modelo basado en agentes, para estudiar el impacto que tiene la cohesión social y la victimización en el proceso de la toma de decisión de un criminal. En el modelado se consideraron los aspectos criminológicos establecidos desde las teorías del desorden social y de las actividades rutinarias en una arquitectura de inteligencia artificial PECS (Physis, Emotion, Cognitive, Social), con el objeto de articular todas las dimensiones que intervienen en la construcción de la decisión. La cohesión social será representada por redes sociales siguiendo el experimento de Watts - Strogatz, y la victimización como resultado acumulado de victimizaciones previas. Los resultados obtenidos de la simulación permitieron replicar las conductas sociales características del actor criminal, ya que en ellos se resaltan la importancia que tiene la percepción de la cohesión social sobre la memoria de victimización generada por eventos criminales pasados, elementos que presentan consistencia con la evidencia empírica de los principios teóricos establecidos en el desorden social y las acciones rutinarias.

1 January 2019 in Revista Tecnura
Verified
Journal Article
Open access
PDF file
GSGuillermo Roberto Solarte
Guillermo Roberto Solarte
JMJosé Soto Mejía
José Soto Mejía
LGLuis Eduardo Muñoz Guerrero
Luis Eduardo Muñoz Guerrero
Contexto: Esta investigación resuelve el problema de encontrar el punto óptimo de localización de una flota de vehículos recolectores de basura y las rutas óptimas para minimizar el costo de su recolección, en 144 barrios del municipio de Dosquebradas, Risaralda (Colombia), utilizando 8 vehículos con capacidad homogénea de 25 toneladas de la empresa Serviciudad. Métodos: Primero, se utilizó una heurística de barrido (Ospina Toro y Orrego, 2016) para encontrar un buen punto de partida para los vehículos de recolección y generar rutas iniciales de buena calidad. Posteriormente, estas rutas iniciales alimentan el algoritmo genético modificado de Chu-Beasley (Solarte, Castillo y Rodríguez, 2015) teniendo en cuenta la capacidad de carga del vehículo (Rondon et al., 2010). Finalmente, para garantizar un resultado óptimo, el mejor encontrado en la fase anterior es tratado nuevamente con una metaheurística tabú (Bodas, 2017). Resultados: Se diseñó una nueva metodología, denominada híbrida CSGTR (Clustering, sweep, genetic, tabu routing) que permitió aprovechar las ventajas de la clusterización (Rueda et al., 2017) antes del ruteo de vehículos (Hernández, 2017), incluyendo modelos heurísticos como la técnica de barrido (Ospina Toro y Orrego, 2016) y metaheurísticos como los algoritmos de Chu-Beasley y tabú (Grajales, Hincapié y Montoya, 2017). La aplicación de la metodología CSGTR permitió reducir el tiempo y los costos de los recorridos de los camiones recolectores de basura en el municipio de Dosquebradas, Risaralda (Colombia). Conclusiones: La metodología hibrida CSGTR para resolver el problema de ubicación de flotas de vehículos y generación de rutas de recolección se presenta como un enfoque alternativo, con mejores resultados que el enfoque previo.

23 September 2026 in Information
Verified
Journal Article
Open access
PDF file
CZCarlos Mauricio Zuluaga-Ramírez
Carlos Mauricio Zuluaga-Ramírez
MGManuela del Pilar Gómez-Suta
Manuela del Pilar Gómez-Suta
MGManuela Idarraga Idarraga Grajales
Manuela Idarraga Idarraga Grajales
CGCarolina Gómez-Carrillo
Carolina Gómez-Carrillo
JCJulio Cesar Chavarro-Porras
Julio Cesar Chavarro-Porras
JSJosé Soto-Mejía
José Soto-Mejía
Ontologies enable the construction of formal and reusable semantic representations of domain knowledge; however, keeping them updated in response to the dynamic evolution of concepts remains a challenge, particularly in the leadership domain, where the diversity of approaches, theories, models, and styles generates conceptual ambiguity and information overlap. To address this limitation, this study proposes an ontology learning methodology based on topic modeling (TM) to create an adaptive component capable of identifying candidate concepts for enriching a previously developed leadership ontology. The methodology is structured into four stages: (1) corpus preprocessing; (2) manual labeling of the corpus evaluated through inter-annotator agreement using the Kappa index; (3) vocabulary construction by comparing five statistical weighting schemes (Inverse Document Frequency (IDF), Entropy, and three variants of Ochoa’s proposal) against a gold standard; and (4) sentence classification and concept construction using four probabilistic topic models (Latent Semantic Analysis (LSA), Non-Negative Matrix Factorization (NMF), Latent Dirichlet Allocation (LDA), and Biterm Topic Model (BTM)), taking as input both a corpus fully associated with the initial ontology and one partially associated. The IDF weighting scheme achieved the best balance between precision and recall (F-measure: 67.03%), and a Support Vector Machine (SVM) classifier reached an accuracy of 98.19% in sentence classification. The LDA and NMF models obtained the highest scores for thematic coherence. However, the expert evaluation revealed that the thematic coherence metric alone is insufficient for evaluating the resulting structures, as it favored the LDA model, while the NMF model generated more conceptually diverse structures. The proposed methodology enables the semi-automatic transformation of unstructured text into structured candidate concepts for subsequent ontology enrichment.