Nowadays, “machine learning” is present in several aspects of the current world, internet advisors, advertisements and “smart” devices that seem to know what we need in a given moment. These are some examples of the problems solved by machine learning.
This book presents the past, the present and the future of the different types of machine learning algorithms. At the beginning of the book, the author takes us to the first years of the computing science, where a programmer had to do absolutely everything by himself to make an algorithm do a certain task. As time passes, there appeared the first...
Frati, Fernando Emmanuel
The main objective of this paper is to propose a software implementation model for concurrency error detection tools that allows reducing process overhead without decreasing its detection capacity. The general model proposed uses software dynamic instrumentation in such a way that an analysis routine can be activated from a signal generated by a hardware event that indicates the possibility of an error occurring. The results obtained showed that, for the case study (an atomicity violation detection algorithm called AVIO), the version that uses the model proposed can detect the same bugs as the original version, but in only 25% of...
Obrist Bertrand, Víctor Udo; Martínez Jara, Eustaquio Alcides
The project is an application that allows users to interact in a virtual environment via a web interface, in which models are in three dimensions to simulate different activities. The application focuses on education with virtual reality technology, enriching the student's perception through the interaction with objects in an artificial world, facilitating their learning. A website was designed for this purpose, on which the application can be executed directly in the web browser with the help of a plugin or downloaded and run as a native application. The user navigates in the virtual environment containing a three-dimensional replica of one...
Huapaya, Constanza Raquel; Guccione, Leonel; Benchoff, Delia Esther; Lizarralde, Francisco Ángel José; Gonzalez, Marcela P.
A module of a Student Model in a Virtual Learning Environment will be presented in order to promote the personalization of instructional material based on the dynamic knowledge levels and learning styles. The improvement is provided by the inclusion experts' experience in the teaching field whose opinions have been expressed in fuzzy rules using two input linguistic variables (knowledge level and learning style) and two linguistic output variables (difficulty level and individual/group work). These last two variables categorize instructional materials.
Aragón, Victoria S.; Esquivel, Susana Cecilia
In this paper, an algorithm inspired on the T- Cell model of the immune system is presented, it is used to solve Economic Dispatch Problems with smooth objective function. The proposed approach is called IA EDP S, which stands for Immune Algorithm for Economic Dispatch Prob- lem for smooth objective function, and it uses as differentiation process a redistribution power op- erator. The proposed approach is validated using five problems taken from the specialized litera- ture. Our results are compared with respect to those obtained by several other approaches.
Wolfmann, Aaron Gustavo Horacio; De Giusti, Armando Eduardo
This paper introduces the PN-PEM framework. It is based on the representation of an algorithm with Petri Nets. Frequently, a real algorithm needs a large Petri Net to be represented. We present a way to model an algorithm with Colored Petri Nets that simplify the model. After that, this high level model is transformed into a low level but executable model, preserving its semantics. The execution also needs other components of the framework, as the involved processors, data used and executable kernels. The combination of these elements is described in order to obtain a parallel execution. Some tests are also...
Garciarena Ucelay, María José; Villegas, María Paula; Cagnina, Leticia; Errecalde, Marcelo Luis
Author Profiling is the task of predicting characteristics of the author of a text, such as age, gender, personality, native language, etc. This is a task of growing importance due to the potential applications in security, crime detection and marketing, among others. An interesting point is to study the robustness of a classifier when it is trained with a data set and tested with others containing different characteristics. Commonly this is called cross domain experimentation.
Although different cross domain studies have been done for data sets in English language, for Spanish it has recently begun. In this context, this work presents...
Artola, Verónica; Sanz, Cecilia Verónica; Moralejo, Lucrecia; Pesado, Patricia Mabel; Baldassarri Santalucía, Sandra
This paper presents the design of an authoring tool for creating educational activities based on tangible interaction. First, background information is presented in relation to tangible interaction and its possibilities for the educational context. Also, the concept of authoring tool is reviewed, and a number of desirable features for authoring tools oriented to the educational context are listed. Finally, TIES (Tangible Interaction in Educational Scenario) is described, which is the authoring tool we designed and upon which this paper is focused. The reasons why this is a necessary tool are discussed, and some conclusions and future lines of work are...
Roggero, Patricia; Reyes, Nora Susana; Figueroa, Karina; Paredes, Rodrigo
Similarity search is a difficult problem and various indexing schemas have been defined to process similarity queries efficiently in many applications, including multimedia databases and other repositories handling complex objects. Metric indices support efficient similarity searches, but most of them are designed for main memory. Thus, they can handle only small datasets, suffering serious performance degradations when the objects reside on disk. Most reallife database applications require indices able to work on secondary memory.
Among a plethora of indices, the List of Clustered Permutations (LCP) has shown to be competitive in main memory.We introduce a secondary-memory variant of the LCP, which...
Baldini, Patricia Noemí; Calandrini, Guillermo; Doñate, Pedro
This paper formulates a robust control for variable time-delay system models. An automatic tuning method for PID-type controller is proposed. The adopted method integrates robust control design using Quantitative Feedback Theory (QFT) with Particle Swan Optimization heuristic algorithms (PSO) to systematize the loop-shaping stage. The objective of the design method is to reach a good compromise among robust stability, robust tracking and disturbance rejection with minimal control effort.
The resulting algorithm has attractive features, such as easy implementation, stable convergence characteristic and good computational efficiency. In particular, the results of the control design for active queue management (AQM) systems are presented.
Lobos, Jair; Gil Costa, Graciela Verónica; Reyes, Nora Susana; Printista, Alicia Marcela; Marín, Mauricio
The classic use of Stream Processing platforms enables working with data in real time, which allows you to generate data analysis quickly attending to a decisionmaking process. However, you can use these platforms for other applications such as indexing and subsequent use of similarity search objects in a database. The images can be displayed on a metric space, which has features that allow rules to discard a not similar image quickly without making costly computations. This paper presents the use of a Stream Processing platform to index images generated by different users. For this, it is necessary to represent these...
Turull Torres, José María
We prove that the existential fragment Σ12,ω of the third order logic TOω captures the relational complexity class non deterministic exponential time. As a Corollary we have that relational machines that work in NEXPTIMEr can simulate third order relational machines that work in NEXPTIME3,r.
Kohan, Mariano; Ale, Juan María
The discovery of moving objects trajectory patterns representing a high traffic density have been covered on different works using diverse approaches. These models are useful for the areas of transportation planning, traffic monitoring and advertising on public roads. Besides of the important utility, these type of patterns usually do not specify a difference between a high traffic and a traffic congestion. In this work, we propose a model for the discovery of high traffic flow patterns and traffic congestions, represented in the same pattern. Also, as a complement, we present a model that discovers alternative paths to the severe traffic...
Anabalon, Diego; Garriga, Martín; Flores, Andrés P.; Cechich, Alejandra; Zunino, Alejandro
Building Service-oriented Applications implies the selection of adequate services to fulfill required functionality. Even a reduced set of candidate services involves an overwhelming assessment effort. In a previous work we have presented an approach to assist developers in the selection of Web Services. In this paper we detail its behavioral assessment procedure, which is based on testing and adaptation. This is done by using black-box testing criteria to explore services behavior. In addition, helpful information takes shape to build the needed adaptation logic to safely integrate the selected candidate into a Service-oriented Application. A concise case study shows the potential...
Gómez, Sergio Alejandro
We present a preliminary framework for reasoning with possibilistic description logics ontologies with disjunctive assertions (PoDLoDA ontologies for short). Given a PoDLoDA ontology, its terminological box is expressed in the description logic programming fragment but its assertional box allows four kinds of statements: an individual is a member of a concept, two individuals are related through a role, an individual is a member of the union of two or more concepts or two individuals are related through the union of two or more roles. Axioms and statements in PoDLoDA ontologies have a numerical certainty degree attached. A disjunctive assertion expresses...
Balladini, Javier; Rozas, Claudia; Frati, Fernando Emmanuel; Vicente, Néstor; Orlandi, Cristina
In a typical intensive care unit of a healthcare facilities, many sensors are connected to patients to measure high frequency physiological data. Currently, measurements are registered from time to time, possibly every hour. With this data lost, we are losing many opportunities to discover new patterns in vital signs that could lead to earlier detection of pathologies. The early detection of pathologies gives physicians the ability to plan and begin treatments sooner or potentially stop the progression of a condition, possibly reducing mortality and costs. The data generated by medical equipment are a Big Data problem with near real-time restrictions...
Aquino, Germán; Lanzarini, Laura Cristina
The large amount of textual information digitally available today gives rise to the need for effective means of indexing, searching and retrieving this information.
Keywords are used to describe briefly and precisely the contents of a textual document. In this paper we present an algorithm for keyword extraction from documents written in Spanish.This algorithm combines autoencoders, which are adequate for highly unbalanced classification problems, with the discriminative power of conventional binary classifiers. In order to improve its performance on larger and more diverse datasets, our algorithm trains several models of each kind through bagging.
Selzer, Matías Nicolás; Larrea, Martín Leonardo
Virtual Reality has become, once again, a popular and interesting topic, both as a research and commercial field. This trend has its origin in the use of mobile devices as computational core and displays for Virtual Reality. Android is one of the most used platform in this context and Unity3d is a suitable game engine for such platform. In order to improve the immersive experience, some electronic devices, Arduino especially, are used to gather information, such as the movement of the user’s arms or legs.
Although Android, Arduino and Unity3d are often used independently in Virtual Reality investigations, few studies use...
De Giusti, Armando Eduardo
Descripción de los contenidos del presente número de la revista.