Estimate of the satisfaction attribute in user tests from facial expression analysis
Darly Mildred Delgado, Diego Fernando Girón Timaná, Gabriel Elías Chanchí, Katerine Márceles...
Usability has become a fundamental aspect of the development of a software product for being a primordial feature of quality. This concept is linked to the use nature of software and can be evaluated through a usability test. This testing is made quantitatively through measurement for two of the features (efficacy and efficiency) that define it. The satisfaction as a third associated feature is of a qualitative kind which estimation is subjective given that it depends on the absence of discomfort by the test user during its interaction with the product. This article is the result of a research that proposes an approach for the estimation of the satisfaction in user tests involving facial expression analysis. This idea is supported in the development of a software system for the emotional monitoring of a user during a test.
QoS Analysis for IPTV in a software defined network environment
Juan Camilo Caviedes Valencia, Wilmar Yesid Campo Muñoz, Gabriel Elías Chanchí Golondrino
The massive deployment of services supported on the video streaming technology brings with itself the expansion of the data networks and the growth of the bandwidth necessary for carrying information. This demands greater control capabilities and network management that guarantees the quality of the service (QoS) for the final user. Such a demand has pushed the innovation of network architectures. Generating thus software-defined networks (SDN), network architecture paradigm where the data and control blueprints are uncoupled. This article analyzes the QoS metrics of a real and emulated SDN network for one of the video streaming services such as the television over the IP protocol (IPTV). The results show the possibility of equaling the QoS of a conventional architecture while offering other advantages for the network such as centralized logic and programmability at the management level.
Modeling and nonlinear control of volatile fatty acids in an upflow anaerobic fixed bed biofilm reactor (UAFBR)
Ivan Dario Ramirez, Andrés Ramírez Villareal, Juan David Ospina Nieto
In this article a robust regulation strategy of volatile fatty acids is proposed by an anaerobic digestion process. The proposed strategy is a control law via nonlinear feedback obtained by input-output linearization that allows the completion of the control goals despite the modeling errors (uncertainties in the kinetics of the reactions) and the actuator restrictions. The experiments were conducted with an pilot fixed anaerobic reactor and ascending flow (UAFBR) with an active volume of 0.947 m3 fully equipped, used for the treatment of residual waters coming from a wine distillery in Narbonne, France. With the goal of validating the models used, simulated and experimental results are presented and discussed. The performance parameters of the controller under the AM2 and ADM1 models (up time, establishment time, over-impulse and stationary state error) obtained from the numeric simulations that prove the efficiency and applicability of the proposed control strategy.
Application of learning analytics and educational data mining in an institution of superior education in Colombia
Johnny Alexander Salazar Cardona, Jorge Iván Triviño Arbelaez
Nowadays, data is a key element for the continuous improvement of an organization’s decision-taking, achieved through the application of awareness and knowledge processes by undergoing a pre-processing, transformation and analysis over the data. The academic field is aware of this kind of application and is a trend for the exploitation of data generated by the students, its management and academics dependencies on a daily basis in order to continuously improve the processes.
Current methodologies propose two different guidelines: Learning Analytics (LA), primarily focused on descriptive processes, and Educational Data Mining (EDM) for predictive processes, directing activities adjusted to this environment for obtaining satisfactory results. It is for this reason that this article presents an application of these two guidelines in a higher education institution, focusing them on sensitive data of the students that will support the high direction decision-taking in these institutions.
Maze companion: artificial intelligence application that supports mazes solving
Yana Elida Saint-Priest, Oscar Domínguez-Acevedo, Carlos Márquez-Hernández
The concentration capabilities of a child depend on his or her age and the normal process is that along with their growth their attention span grows as well; nonetheless, it is important to work on this capability from a young age. The classic board games such as chess, mazes, triqui, puzzles, etc. require a certain amount of mental processes in order to be played and are ideal for keeping the attention of the child for several minutes. This project is an innovative proposal through image processing and width search algorithms; it solves physical mazes created by the players. The obtained effectiveness was 72 %, remarking that the quality of the sensor and the angle in which the photograph is captured is of enormous influence.
Systematic review of literature on electrical energy consumption forecast models
Agustín José Mazzeo, Lina María Sepúlveda Cano, Luisa Fernanda Villa Montoya, Ricardo Alonso...
The growing consumption of electrical energy, climate change and the development of new technologies demand improvements for efficient energy management. An adequate forecast of the energy consumption is relevant for the sustainable development of any country. This article proposes a systematic review of selected literature based on search chains formed by the terms forecasting, energy and consumption applied to the scientific databases. In the article are compared mostly the models/ techniques used, the considered variables and the error metrics used for obtaining knowledge on each one of the proposals, relieve its features and thus highlight the void in the literature that might be determinant for new research work. As conclusions are made evident the continuous use of neural networks for forecasting the energy consumption, the importance of determining the input variables and the error measuring for evaluating the precision of the models. Finally, the development of a model for the CEE short term forecast of a Latin-American developing country based on the comparison and evaluation of different techniques/models, variables and already existing tools is proposed as a new line of research.
This article proposes a framework for the implementation of Lean Laboratories in higher education laboratories, which was applied in the soil and pavements laboratory of the Escuela Colombiana de Ingeniería Julio Garavito. With this framework, which incorporates in a detailed manner the steps for its application, the project aims towards decreasing the residues in these environments such as insufficient training, lack of supervision and experience, systematic performance measures, responsibility, materials availability, and generally, the absence of a systematic disposition of the laboratory and the equipment. The results achieved show important improvements in the short and medium-term in the registry and lending processes of tools and equipment. The analyzed case displays positive results with the potential of being exemplary for other organizations in the sector with aims towards implementation Lean and its tools as a functional, practical, and low-cost alternative.
Characterization of the In-Phase Response and Phase Compensation in Multicarrier Systems for Channels with Non-Uniform Frequent
Claudia Lucía Cortés Cortés, Neil Guerrero González
This work proposes a methodology for the phase estimation in sub-bands of a non-uniform response channel. The novelty of this proposal lies in the use of a modulation format in QSPK (Quadrature Phase Shift Keying) quadrature-phase modified as a test signal for the phase estimation that eliminates the ambiguity issue. The methodology here proposed is verified experimentally using a defective RF cable as transmission channel and thus achieving data transmission rates of 20 Kbaud in a 120 cm distance in a frequency band between the 50 kHz y 276 kHz.
A Literature Review of Optimal Generation Start-Up Methodologies for Power System Restoration considering Conventional and Non-Conventional Renewable Energy Sources
Ricardo Andrés Pardo, Jesus Maria Lopez-Lezama
After a large-scale blackout power system restoration must be accomplished as soon as possible. For this, the generation must be initially restored, then the transmission system, and finally the load pick up must be completed. To obtain a faster restoration process, it is necessary to establish start-up methodologies for generating units that first start those units that provide black start, and then take the starting power to other generation units without this characteristic, by means of a feasible transmission route. This paper presents a review of different methodologies of optimal generation start-up for power system restoration reported in the scientific literature taking into account the integration of non-conventional renewable energy sources. Within this review it is highlighted that heuristic methods, despite of being effective, are not used in real-time operation due to their high computational cost.
Sustainability Indicators of the Amazonian Thermoelectrics: the Rondônia’s Case Study
Cristiano Torres do Amaral, Artur de Souza Moret, José Manuel Carvalho Marta
This article presents a methodology for the analysis of the sustainability of thermoelectric plants with a highlight on the isolated system in the Amazon, in the state of Rondônia, Brasil. The document describes the indicators for the constructions of a general indicator of sustainability that allows the classification of micro-networks with thermoelectrics according to their environmental, social and, economic contributions. In this study we use official data and polls made to the electric distribution companies. The model proposed in this paper contributes to the definition of policies for the control of greenhouse gases, mitigation of risks, new technologies development and, the distribution of public resources for alternative energy sources research. This research project was developed by the Research Group on Renewable and Sustainable Energy of the Universidade Federal de Rondônia.
Ablative Composites of Polymeric Matrix (Polyester Resin) Reinforced With Industrial Residue Materials and Fiberglass
Julián Portocarrero Hermann, John Darío Osorio, Rafael Andres Robayo-Salazar
Ablative materials are used in thermal protection systems to protect equipment or elements that are exposed to high temperatures in service. Generally, these materials are difficult to acquire and high cost, which hinders their application in countries such as Colombia. This research describes the obtaining and characterization of ablative composite materials based on a polyester resin matrix reinforced with particulate materials (60-70% by weight) and short glass fiber (5% by weight). Ceramic powder and finely ground aluminum slag were used as particulates (˂75 μm), highlighting that both are industrial waste or by-products. The developed composites were subjected to the direct flame test in order to evaluate their ablative performance, simulating the real extreme conditions (1600-2000 °C) of the nozzle of a rocket (probe type). In addition, the effect of the ablative test on the level of cracking and erosion (weight loss) of the composites developed was evaluated. As an indirect measure of the residual mechanical performance of the ablative composites, the hardness of the thermally affected area was determined. The results obtained are considered of great importance for the production of thermal protection systems and propulsion components with low cost and available raw materials in Colombia.
The study of the motivation of the future software engineers is of the uttermost importance taking into account the high demand unsatisfied by these professionals and the responsibility of the academy of promoting for a significative and active learning-teaching process that achieves positively compromised students. The contemplated objectives were: a) identifying the aspects related in the motivation of informatics students with an emphasis in the software engineering area; b) ascertaining the reliability and validity of the elaborated instrument; c) identifying the motivation levels of the studied population. For achieving these objectives, a new instrument was designed and validated, a two-phase data analysis was also performed for obtaining a factors and subfactors structure that gives account of the motivational level. In the first phase of the analysis we obtained a four factor statistic model (performance attributes, relationship between students and teacher; performance of the teachers; physical and virtual resources) that explain the 71,5 % of variability. In the second phase of analysis the model for identifying the factors was refined and with that a high level of reliability was achieved. Regarding the performance attributes, we evidences a necessity for helping students to handle their own time; the pedagogic strategies display a high predictive power on the motivation of the students in the software engineering area. This article concludes that the factors and subfactors structures structures obtained with a high level of reliability served as a starting point for identifying improvement actions in the institutions responsible for the writing of the articles; the reliability and validity of this instrument profile it as a candidate for being applied in similar programs for other institutions.