GREC  Grup de Recerca en Enginyeria del Coneixement
http://hdl.handle.net/2117/3350
20171218T13:06:29Z

Handling binary classification problems with a priority class by using Support Vector Machines
http://hdl.handle.net/2117/111896
Handling binary classification problems with a priority class by using Support Vector Machines
Gonzalez Abril, Luis; Angulo Bahón, Cecilio; Núñez Castro, Haydemar; Leal, Yenny
© 2017 Elsevier B.V. A postprocessing technique for Support Vector Machine (SVM) algorithms for binary classification problems is introduced in order to obtain adequate accuracy on a priority class (labelled as a positive class). That is, the true positive rate (or recall or sensitivity) is prioritized over the accuracy of the overall classifier. Hence, false negative (or Type I) errors receive greater consideration than false positive (Type II) errors during the construction of the model. This postprocessing technique tunes the initial bias term once a solution vector is learned by using standard SVM algorithms in two steps: First, a fixed threshold is given as a lower bound for the recall measure; second, the true negative rate (or specificity) is maximized. Experiments, carried out on eleven standard UCI datasets, show that the modified SVM satisfies the aims for which it has been designed. Furthermore, results are comparable or better than those obtained when other stateoftheart SVM algorithms and other usual metrics are considered.
20171213T12:30:37Z
Gonzalez Abril, Luis
Angulo Bahón, Cecilio
Núñez Castro, Haydemar
Leal, Yenny
© 2017 Elsevier B.V. A postprocessing technique for Support Vector Machine (SVM) algorithms for binary classification problems is introduced in order to obtain adequate accuracy on a priority class (labelled as a positive class). That is, the true positive rate (or recall or sensitivity) is prioritized over the accuracy of the overall classifier. Hence, false negative (or Type I) errors receive greater consideration than false positive (Type II) errors during the construction of the model. This postprocessing technique tunes the initial bias term once a solution vector is learned by using standard SVM algorithms in two steps: First, a fixed threshold is given as a lower bound for the recall measure; second, the true negative rate (or specificity) is maximized. Experiments, carried out on eleven standard UCI datasets, show that the modified SVM satisfies the aims for which it has been designed. Furthermore, results are comparable or better than those obtained when other stateoftheart SVM algorithms and other usual metrics are considered.

Multivariate Regression with Incremental Learning of Gaussian Mixture Models
http://hdl.handle.net/2117/110920
Multivariate Regression with Incremental Learning of Gaussian Mixture Models
Acevedo Valle, Juan Manuel; Trejo Ramírez, Karla Andrea; Angulo Bahón, Cecilio
Within the machine learning framework, incremental learning of multivariate spaces is of special interest for online applications. In this work, the regression problem for multivariate systems is solved by implementing an efficient probabilistic incremental algorithm. It allows learning highdimensional redundant nonlinear maps by the cumulative acquisition of data from inputoutput systems. The proposed model is aimed at solving prediction and inference problems. The implementation introduced in this work allows learning from data batches without the need of keeping them in memory afterwards. The learning architecture is built using Incremental Gaussian Mixture Models. The ExpectationMaximization algorithm and general geometric properties of Gaussian distributions are used to train the models. Our current implementation can produce accurate results fitting models in real multivariate systems. Results are shown from testing the algorithm for both situations, one where the incremental learning is demonstrated and the second where the performance to solve the regression problem is evaluated on a toy example.
La publicació definitiva d'aquest treball està disponible a IOS Press a través de http://dx.doi.org/10.3233/9781614998068196
20171120T14:26:47Z
Acevedo Valle, Juan Manuel
Trejo Ramírez, Karla Andrea
Angulo Bahón, Cecilio
Within the machine learning framework, incremental learning of multivariate spaces is of special interest for online applications. In this work, the regression problem for multivariate systems is solved by implementing an efficient probabilistic incremental algorithm. It allows learning highdimensional redundant nonlinear maps by the cumulative acquisition of data from inputoutput systems. The proposed model is aimed at solving prediction and inference problems. The implementation introduced in this work allows learning from data batches without the need of keeping them in memory afterwards. The learning architecture is built using Incremental Gaussian Mixture Models. The ExpectationMaximization algorithm and general geometric properties of Gaussian distributions are used to train the models. Our current implementation can produce accurate results fitting models in real multivariate systems. Results are shown from testing the algorithm for both situations, one where the incremental learning is demonstrated and the second where the performance to solve the regression problem is evaluated on a toy example.

A Transformational creativity tool to support musical composition
http://hdl.handle.net/2117/110739
A Transformational creativity tool to support musical composition
Raya Giner, Cristóbal; Ruiz Vegas, Francisco Javier; Angulo Bahón, Cecilio; Samà Monsonís, Albert; Agell, Nuria
In this paper we use the idea of conceptual space introduced by Boden
and redefine some properties such appropriateness and relevance that facilitate the
computational implementation of the transformational creativity mechanism. While
appropriateness can only be evaluated by an expert, relevance can be objectively
measured for any spectator. Computational creativity is based on the relationship
between appropriateness and relevance of a concept, and therefore a computational
system can be used to support this task. The paper analyses this relationship in the
field of music in order to obtain a computer tool to support the musical composition
task.
20171116T10:58:21Z
Raya Giner, Cristóbal
Ruiz Vegas, Francisco Javier
Angulo Bahón, Cecilio
Samà Monsonís, Albert
Agell, Nuria
In this paper we use the idea of conceptual space introduced by Boden
and redefine some properties such appropriateness and relevance that facilitate the
computational implementation of the transformational creativity mechanism. While
appropriateness can only be evaluated by an expert, relevance can be objectively
measured for any spectator. Computational creativity is based on the relationship
between appropriateness and relevance of a concept, and therefore a computational
system can be used to support this task. The paper analyses this relationship in the
field of music in order to obtain a computer tool to support the musical composition
task.

Improving SVM classification on imbalanced datasets by introducing a new bias
http://hdl.handle.net/2117/110122
Improving SVM classification on imbalanced datasets by introducing a new bias
Núñez Castro, Haydemar; Gonzalez Abril, Luis; Angulo Bahón, Cecilio
Support Vector Machine (SVM) learning from imbalanced datasets, as well as most learning machines, can show poor performance on the minority class because SVMs were designed to induce a model based on the overall error. To improve their performance in these kind of problems, a lowcost postprocessing strategy is proposed based on calculating a new bias to adjust the function learned by the SVM. The proposed bias will consider the proportional size between classes in order to improve performance on the minority class. This solution avoids not only introducing and tuning new parameters, but also modifying the standard optimization problem for SVM training. Experimental results on 34 datasets, with different degrees of imbalance, show that the proposed method actually improves the classification on imbalanced datasets, by using standardized error measures based on sensitivity and gmeans. Furthermore, its performance is comparable to wellknown costsensitive and Synthetic Minority Oversampling Technique (SMOTE) schemes, without adding complexity or computational costs.
20171108T06:51:29Z
Núñez Castro, Haydemar
Gonzalez Abril, Luis
Angulo Bahón, Cecilio
Support Vector Machine (SVM) learning from imbalanced datasets, as well as most learning machines, can show poor performance on the minority class because SVMs were designed to induce a model based on the overall error. To improve their performance in these kind of problems, a lowcost postprocessing strategy is proposed based on calculating a new bias to adjust the function learned by the SVM. The proposed bias will consider the proportional size between classes in order to improve performance on the minority class. This solution avoids not only introducing and tuning new parameters, but also modifying the standard optimization problem for SVM training. Experimental results on 34 datasets, with different degrees of imbalance, show that the proposed method actually improves the classification on imbalanced datasets, by using standardized error measures based on sensitivity and gmeans. Furthermore, its performance is comparable to wellknown costsensitive and Synthetic Minority Oversampling Technique (SMOTE) schemes, without adding complexity or computational costs.

Guest editorial: sensorimotor contingencies for cognitive robotics
http://hdl.handle.net/2117/108957
Guest editorial: sensorimotor contingencies for cognitive robotics
Alenyà Ribas, Guillem; Tellez Lara, Ricardo; O'Regan, Kevin; Angulo Bahón, Cecilio
The sensorimotor approach to cognition states, that the key to bring semantics to the world of a robot, requires making the robot learn the relation between the actions that the robot performs and the change it experiences in its sensed data because of those actions. Those relations are called sensorimotor contingencies (SMCs). This special issue presents a variety of recent developments in SMCs with a particular focus on cognitive robotics applications.
© 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
20171023T08:00:11Z
Alenyà Ribas, Guillem
Tellez Lara, Ricardo
O'Regan, Kevin
Angulo Bahón, Cecilio
The sensorimotor approach to cognition states, that the key to bring semantics to the world of a robot, requires making the robot learn the relation between the actions that the robot performs and the change it experiences in its sensed data because of those actions. Those relations are called sensorimotor contingencies (SMCs). This special issue presents a variety of recent developments in SMCs with a particular focus on cognitive robotics applications.

A decision support tool using order weighted averaging for conference review assignment
http://hdl.handle.net/2117/108657
A decision support tool using order weighted averaging for conference review assignment
Nguyen, Jennifer; Sánchez Hernández, Germán; Agell Jané, Núria; Rovira Llobera, Xari; Angulo Bahón, Cecilio
Assigning papers to reviewers is a large, long and difficult task for conference chairs and scientific committees. The paper reviewer assignment problem is a multiagent problem which requires understanding reviewer expertise and paper topics for the matching process. This paper proposes to elaborate on some features used to compute reviewer expertise and aggregate multiple factors to find the fittest combination of reviewers for each paper. Expertise information is gathered implicitly from publicly available information and a reviewer profile is generated automatically. An Ordered Weighted Averaging (OWA) aggregation function is used to summarize information coming from different sources and rank candidate reviewers for each paper. General constraints for the Reviewer Assignment Problem (RAP) have been incorporated into a real case example: (i) conflicts of interest between a reviewer and authors should be avoided, (ii) each paper must have a minimum number of reviewers, and (iii) each reviewer load cannot exceed a certain number of papers.
20171011T13:00:46Z
Nguyen, Jennifer
Sánchez Hernández, Germán
Agell Jané, Núria
Rovira Llobera, Xari
Angulo Bahón, Cecilio
Assigning papers to reviewers is a large, long and difficult task for conference chairs and scientific committees. The paper reviewer assignment problem is a multiagent problem which requires understanding reviewer expertise and paper topics for the matching process. This paper proposes to elaborate on some features used to compute reviewer expertise and aggregate multiple factors to find the fittest combination of reviewers for each paper. Expertise information is gathered implicitly from publicly available information and a reviewer profile is generated automatically. An Ordered Weighted Averaging (OWA) aggregation function is used to summarize information coming from different sources and rank candidate reviewers for each paper. General constraints for the Reviewer Assignment Problem (RAP) have been incorporated into a real case example: (i) conflicts of interest between a reviewer and authors should be avoided, (ii) each paper must have a minimum number of reviewers, and (iii) each reviewer load cannot exceed a certain number of papers.

Evaluating studentinternship fit by using fuzzy linguistic terms and a fuzzy OWA operator
http://hdl.handle.net/2117/107585
Evaluating studentinternship fit by using fuzzy linguistic terms and a fuzzy OWA operator
Jennifer, Nguyen; Sánchez Hernández, Germán; Armisen Morell, Albert; Agell, Nuria; Angulo Bahón, Cecilio
Personnel selection is a wellknown problem that is made difficult by incomplete and imprecise information about candidate and position compatibility. This paper shows how positions, which satisfy candidate’s interests, can be identified with fuzzy linguistic terms and a fuzzy OWA operator. A set of relevant positions aligned with a student’s interests is selected using this approach. The mplementation of the proposed method is illustrated using a numerical example in a business application.
© 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
20170913T09:25:16Z
Jennifer, Nguyen
Sánchez Hernández, Germán
Armisen Morell, Albert
Agell, Nuria
Angulo Bahón, Cecilio
Personnel selection is a wellknown problem that is made difficult by incomplete and imprecise information about candidate and position compatibility. This paper shows how positions, which satisfy candidate’s interests, can be identified with fuzzy linguistic terms and a fuzzy OWA operator. A set of relevant positions aligned with a student’s interests is selected using this approach. The mplementation of the proposed method is illustrated using a numerical example in a business application.

A qualitative spatial descriptor of grouprobot interactions
http://hdl.handle.net/2117/107581
A qualitative spatial descriptor of grouprobot interactions
Falomir, Zoe; Angulo Bahón, Cecilio
The problem of finding a suitable qualitative representation for robots to reason about activity spaces where they carry out tasks such as leading or interacting with a group of people is tackled in this paper. For that, a Qualitative Spatial model for Group Robot Interaction (QSGRI) is proposed to define Kendon’s Fformations [16] depending on: (i) the relative location of the robot with respect to other individuals involved in that interaction; (ii) the individuals’ orientation; (iii) the shared peripersonal distance; and (iv) the role of the individuals (observer, main character or interactive). An iconic representation is provided and Kendon’s formations are defined logically. The conceptual neighborhood of the evolution of Kendon formations is studied, that is, how one formation is transformed into another. These transformations can depend on the role that the robot have, and on the amount of people involved.
20170913T08:27:20Z
Falomir, Zoe
Angulo Bahón, Cecilio
The problem of finding a suitable qualitative representation for robots to reason about activity spaces where they carry out tasks such as leading or interacting with a group of people is tackled in this paper. For that, a Qualitative Spatial model for Group Robot Interaction (QSGRI) is proposed to define Kendon’s Fformations [16] depending on: (i) the relative location of the robot with respect to other individuals involved in that interaction; (ii) the individuals’ orientation; (iii) the shared peripersonal distance; and (iv) the role of the individuals (observer, main character or interactive). An iconic representation is provided and Kendon’s formations are defined logically. The conceptual neighborhood of the evolution of Kendon formations is studied, that is, how one formation is transformed into another. These transformations can depend on the role that the robot have, and on the amount of people involved.

Obstacle avoidance basedvisual navigation for micro aerial vehicles
http://hdl.handle.net/2117/104325
Obstacle avoidance basedvisual navigation for micro aerial vehicles
Aguilar Castillo, Wilbert Geovanny; Casaliglla, Veronica P.; Polit, Jose L.
This paper describes an obstacle avoidance system for lowcost Unmanned Aerial Vehicles (UAVs) using vision as the principal source of information through the monocular onboard camera. For detecting obstacles, the proposed system compares the image obtained in real time from the UAV with a database of obstacles that must be avoided. In our proposal, we include the feature point detector Speeded Up Robust Features (SURF) for fast obstacle detection and a control law to avoid them. Furthermore, our research includes a path recovery algorithm. Our method is attractive for compact MAVs in which other sensors will not be implemented. The system was tested in real time on a Micro Aerial Vehicle (MAV), to detect and avoid obstacles in an unknown controlled environment; we compared our approach with related works.
20170511T15:01:56Z
Aguilar Castillo, Wilbert Geovanny
Casaliglla, Veronica P.
Polit, Jose L.
This paper describes an obstacle avoidance system for lowcost Unmanned Aerial Vehicles (UAVs) using vision as the principal source of information through the monocular onboard camera. For detecting obstacles, the proposed system compares the image obtained in real time from the UAV with a database of obstacles that must be avoided. In our proposal, we include the feature point detector Speeded Up Robust Features (SURF) for fast obstacle detection and a control law to avoid them. Furthermore, our research includes a path recovery algorithm. Our method is attractive for compact MAVs in which other sensors will not be implemented. The system was tested in real time on a Micro Aerial Vehicle (MAV), to detect and avoid obstacles in an unknown controlled environment; we compared our approach with related works.

3D environment mapping using the Kinect V2 and path planning based on RRT algorithms
http://hdl.handle.net/2117/103909
3D environment mapping using the Kinect V2 and path planning based on RRT algorithms
Aguilar Castillo, Wilbert Geovanny; Morales, Stephanie G.
This paper describes a 3D path planning system that is able to provide a solution trajectory for the automatic control of a robot. The proposed system uses a point cloud obtained from the robot workspace, with a Kinect V2 sensor to identify the interest regions and the obstacles of the environment. Our proposal includes a collisionfree path planner based on the Rapidlyexploring Random Trees variant (RRT*), for a safe and optimal navigation of robots in 3D spaces. Results on RGBD segmentation and recognition, point cloud processing, and comparisons between different RRT* algorithms, are presented.
20170502T13:30:27Z
Aguilar Castillo, Wilbert Geovanny
Morales, Stephanie G.
This paper describes a 3D path planning system that is able to provide a solution trajectory for the automatic control of a robot. The proposed system uses a point cloud obtained from the robot workspace, with a Kinect V2 sensor to identify the interest regions and the obstacles of the environment. Our proposal includes a collisionfree path planner based on the Rapidlyexploring Random Trees variant (RRT*), for a safe and optimal navigation of robots in 3D spaces. Results on RGBD segmentation and recognition, point cloud processing, and comparisons between different RRT* algorithms, are presented.