Email: andresarciniegas@udenar.edu.co

Research Interests

Supervised learning, smart grids, metering, robotics

Publications

1. Diseño e implementación de un Sistema de Medición Inteligente para AMI de la microrred de la Universidad de Nariño
   DOI: 10.29019/enfoqueute.v8n1.136
   Andrés Fernando Arciniegas Mejía, David Imbajoa, Javier Revelo F.
   See abstract | see full paper
   ×

   Abstract

   Este proyecto describe la propuesta de un Sistema de Gestión de Medida desde el proceso de diseño hasta la implementación de un prototipo de medidor inteligente, para operación en la microrred de la Universidad de Nariño. Este dispositivo se caracteriza por el constante registro de medidas con resolución de 24 bits, suficiente para realizar análisis de calidad de la potencia. Además, se integra al sistema SCADA, a través de los protocolos abiertos MODBUS, DNP3 e IEC 61850 con técnicas de ciberseguridad programada. De igual manera, se diseña varias plataformas de gestión para el monitoreo de variables eléctricas y el control del flujo de corriente a las cargas. Finalmente, dos casos de estudios muestran la aplicación en la microrred, del Sistema de Gestión de Medida implementado.

2. Multi-labeler Classification Using Kernel Representations and Mixture of Classifiers
   Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications: 21st Iberoamerican Congress, CIARP 2016, Lima, Peru, November 8–11, 2016, Proceedings
   David Imbajoa, Iván Gustin, Mauricio Bolaños, Andrés Fernando Arciniegas Mejía, Fredy Guasmayan, María JANETH Bravo Montenegro, A. E. Castro-Ospina, Diego Peluffo
   See abstract | see full paper
   ×

   Abstract

   This work introduces a multi-labeler kernel a novel approach for data classification learning from multiple labelers. The learning process is done by training support-vector machine classifiers using the set of labelers (one labeler per classifier). The objective functions representing the boundary decision of each classifier are mixed by means of a linear combination. Followed from a variable relevance, the weighting factors are calculated regarding kernel matrices representing each labeler. To do so, a so-called supervised kernel function is also introduced , which is used to construct kernel matrices. Our multi-labeler method reach very good results being a suitable alternative to conventional approaches.2

3. Multi-Labeller classification Method based on Mixture of Classifiers and Genetic Algorithm Optimization
   INCISCOS
   Andrés Fernando Arciniegas Mejía, David Imbajoa, Iván Gustin, Mauricio Bolaños, Dario F. Fajardo F., Fredy Guasmayan, María JANETH Bravo Montenegro, Diego Peluffo
   See abstract | see full paper
   ×

   Abstract

   This work presents a new method proposal applied to Multi-Labelers scenarios. This is a situation where labelling individuals in a set of data based on certain characteristics in the process of determining labels to individuals in a set of data based on certain characteristics. Our approach consists in processing a Support Vector Machine classifier to each labelers substantiated on his answers. We formulate a genetic algorithm optimization to obtain a set of weights according to their opinion, in order to penalize each panelist. Finally, their resulting mappings are mixed, and a final classifier is generated, showing to be better than majority vote. For experiments, the well-known Iris database is handled, with multiple simulated artificial labels. The proposed method reaches very good results compared to conventional multi-labeler methods, able to assess the concordance among panelists considering the structure data.