http://hdl.handle.net/2117/3350 2013-05-22T07:54:45Z http://hdl.handle.net/2117/19266 Title: Ranking and selection of unsupervised learning marketing segmentation Authors: Sanchez Hernandez, German; Chiclana, Francisco; Agell Jané, Núria; Aguado Chao, Juan Carlos Abstract: This paper addresses the problem of choosing the most appropriate classification from a given set of classifications of a set of patterns. This is a relevant topic on unsupervised systems and clustering analysis because different classifications can in general be obtained from the same data set. The provided methodology is based on five fuzzy criteria which are aggregated using an Ordered Weighted Averaging (OWA) operator. To this end, a novel multi-criteria decision making (MCDM) system is defined, which assesses the degree up to which each criterion is met by all classifications. The corresponding single evaluations are then proposed to be aggregated into a collective one by means of an OWA operator guided by a fuzzy linguistic quantifier, which is used to implement the concept of fuzzy majority in the selection process. This new methodology is applied to a real marketing case based on a business to business (B2B) environment to help marketing experts during the segmentation process. As a result, a segmentation containing three segments consisting of 35, 98 and 127 points of sale respectively is selected to be the most suitable to endorse marketing strategies of the firm. Finally, an analysis of the managerial implications of the proposed methodology solution is provided. 2013-05-15T16:10:37Z http://hdl.handle.net/2117/19156 Title: A study on output normalization in multiclass SVMs Authors: González Abril, Luis; Velasco Morente, Francisco; Angulo Bahón, Cecilio; Ortega Ramírez, Juan Antonio Abstract: The use of binary support vector machines (SVMs) in multi-classification is addressed in this paper. Margins associated to the bi-classifiers, since they depend on the geometrical disposition of the classes being separated, are, in general, of various magnitudes. In order to overcome this scaling problem, a normalization process should be applied on the SVMs’ outputs. Thus, a new normalization approach is presented based on the convex hulls that contain the classes to be separated. Furthermore, a theoretical study is developed which justifies the proposed approach, and an interpretation is provided. An empirical study is also carried out to compare this normalization with others found in the literature. 2013-05-10T10:50:07Z http://hdl.handle.net/2117/18359 Title: Pain and anxiety treatment based on social robot interaction with children to improve patient experience. Ongoing research Authors: Angulo Bahón, Cecilio; Garriga Berga, Carles; Luaces, Carlos; Perez Payarols, Jaume; Albo-Canals, Jordi; Díaz Boladeras, Marta Abstract: A major focus for children’s quality of life programs in hospitals is improving their experiences during procedures. In anticipation of treatment, children may become anxious and during procedures pain appears. The aim of this article is to introduce a proposal to design pioneering techniques based on the use of social robots to improve the patient experience by eliminating or minimizing pain and anxiety. According to this proposed challenge, this research aims to design and develop specific human-social robot interaction with pet robots. Robot interactive behavior will be designed based on modular skills using soft-computing paradigms. 2013-03-18T08:15:50Z http://hdl.handle.net/2117/18239 Title: External force estimation for textile grasp detection Authors: Colomé Figueras, Adrià; Pardo Ayala, Diego Esteban; Alenyà Ribas, Guillem; Torras, Carme Abstract: Our current work on external force estimation without end-effector force sensor is resented.To verify if a grasp of a textile has been successful, the external wrench applied on the robot is computed online, with a state observer based on a LWPR [3] model of a task. 2013-03-12T17:45:58Z http://hdl.handle.net/2117/17652 Title: AoL: Action Learning: A methodology to capture expertise in adjustment tasks Authors: Ruiz Vegas, Francisco Javier; Samà Monsonís, Albert; Raya Giner, Cristóbal; Agell Jané, Núria Abstract: It is well known that some people can perform a task with greater precision and accuracy than others: they are experts. In the past, experts were interviewed to find out why they have this expertise, but this was not always completely effective because often experts "don't know what they know". In this paper we propose a model of the process of making decisions performed by experts in the final adjustment of products task. Based on this model, we also propose a system based on a machine learning module that facilitates the capture of these expert skills. We give an example to illustrate the process proposed. 2013-02-12T12:10:49Z http://hdl.handle.net/2117/17450 Title: Dynamically consistent probabilistic model for robot motion learning Authors: Pardo Ayala, Diego Esteban; Rozo Castañeda, Leonel; Alenyà Ribas, Guillem; Torras, Carme Abstract: This work presents a probabilistic model for learning robot tasks from human demonstrations using kinesthetic teaching. The difference with respect to previous works is that a complete state of the robot is used to obtain a consistent representation of the dynamics of the task. The learning framework is based on hidden Markov models and Gaussian mixture regression, used for coding and reproducing the skills. Benefits of the proposed approach are shown in the execution of a simple self-crossing trajectory by a 7-DoF manipulator. 2013-01-21T17:15:49Z http://hdl.handle.net/2117/17153 Title: Human - Humanoid Robot Interaction: The 20Q Game Authors: Pardo Ayala, Diego Esteban; Franco Genís, Oscar; Sàez Pons, Joan; Andrés, Amara; Angulo Bahón, Cecilio Abstract: This article introduces the development of a human-robot interaction’s system. The NAO humanoid robot has been programmed to play the popular 20 questions (20Q) game. The system includes modules for both, speech recognition and text-to-speech as mechanisms for exchanging information with the humanoid robot. The system has been tested with non-expert volunteers to analyse which kind of interaction is obtained. 2012-12-19T12:27:50Z http://hdl.handle.net/2117/17152 Title: Uniform sampling of rotations for discrete and continuous learning of 2D shape models Authors: Perez-Sala, Xavier; Igual, Laura; Escalera, Sergio; Angulo Bahón, Cecilio Abstract: Different methodologies of uniform sampling over the rotation group, SO(3), for building unbiased 2D shape models from 3D objects are introduced and reviewed in this chapter. State-of-the-art non uniform sampling approaches are discussed, and uniform sampling methods using Euler angles and quaternions are introduced. Moreover, since presented work is oriented to model building applications, it is not limited to general discrete methods to obtain uniform 3D rotations, but also from a continuous point of view in the case of Procrustes Analysis. 2012-12-19T11:23:07Z http://hdl.handle.net/2117/17120 Title: The Impact of Introducing Therapeutic Robots in Hospital's Organization Authors: Chung Chang, Cecilia Yu; Díaz Boladeras, Marta; Angulo Bahón, Cecilio Abstract: The introduction of robots in health and wellbeing services could mean a great improvement in patients’ life, but also implies a huge change in hospital’s organization. This work introduces a empirical study that investigates the factors influencing the process of adoption of robots in hospitals. Interviews have been completed in a health organization leading this change. Learned lessons lead to define a balance of facilitators and barriers that should help to apply the innovation. This process has been completed from Kotter’s leading change approach, since it would be the roadmap for the whole procedure. 2012-12-12T13:11:40Z http://hdl.handle.net/2117/16808 Title: Survey on 2D and 3D human pose recovery Authors: Perez Sala, Xavier; Escalera Guerrero, Sergio; Angulo Bahón, Cecilio Abstract: Human Pose Recovery approaches have been studied in the eld of Computer Vision for the last 40 years. Several approaches have been reported, and signi cant improvements have been obtained in both data representation and model design. However, the problem of Human Pose Recovery in uncontrolled environments is far from being solved. In this paper, we de ne a global taxonomy to group the model based methods and discuss their main advantages and drawbacks. 2012-10-29T11:03:42Z http://hdl.handle.net/2117/16807 Title: Máquinas de soporte vectorial sobre conjuntos de datos no balanceados: propuesta de un nuevo sesgo Authors: Núñez Castro, Haydemar; González Abril, Luis; Angulo Bahón, Cecilio Abstract: En el aprendizaje con conjuntos de datos no balanceados, la máquina de soporte vectorial (SVM) puede exhibir un bajo rendimiento sobre la clase minoritaria ya que, como otras máquinas de aprendizaje, están diseñadas para inducir un modelo de clasificación basado en un error global. Con el fin de mejorar su desempeño en este tipo de problemas, en este trabajo se propone una estrategia de post-procesamiento basada en el cálculo de un nuevo sesgo o umbral que toma en cuenta la proporción de las clases en el conjunto de datos y que permite ajustar la función aprendida por la SVM para mejorar su desempeño sobre la clase minoritaria. Esta solución no supone la entonación de nuevos parámetros ni la modificación del problema de optimización estándar para entrenar la SVM. Los resultados obtenidos de la experimentación sobre 23 conjuntos de datos con diferentes grados de desbalance, muestran que efectivamente se logra mejorar las clasificaciones sobre la clase minoritaria, medidas en función de g-media y la sensibilidad. 2012-10-29T10:10:20Z http://hdl.handle.net/2117/16758 Title: Experiencia interuniversitaria en el diseño de sistemas interactivos automatizados Authors: Chacón, Alejandro; Ponsa Asensio, Pere; Pérez, Àlex; Vilanova, Ramon Abstract: Este trabajo presenta una experiencia colaborativa en el ambito de la aplicaci6n del diseno de sistemas interactivos automatizados. EI seguimiento de directrices de diseno de interfaces ha perrnitido disminuir la complejidad de la interfaz de supervisi6n y facilitar el acceso a la gesti6n de las alarmas del sistema. 2012-10-19T10:46:37Z http://hdl.handle.net/2117/16750 Title: A study of the human-system interface complexity sources in wastewater treatment plants Authors: Ponsa Asensio, Pere; Pérez, Àlex; Gámiz Caro, Javier Francisco; Manresa-Yee, Cristina; Vilanova, Ramon Abstract: In human-automation interaction it is necessary to define methods and tools to assess the human-system interface complexity. In a first assessment, the use of an evaluation questionnaire aiming at detecting complexity sources in the human supervisory control room tasks in wastewater treatment plants (WWTP) is presented in this paper. The first results show that the elicitation of complexity sources can be useful to improve the interaction between humans and safety-critical systems in the industrial domain. 2012-10-18T09:53:19Z http://hdl.handle.net/2117/16669 Title: Information technologies based questionnaire tool in human-automation systems Authors: Manresa-Yee, Cristina; Ponsa Asensio, Pere; Vilanova, Ramon; Pérez, Àlex Abstract: In this paper we present an Information Technologies (IT) based questionnaire tool in order to support evaluation activities related to industrial automation. We assess the questionnaire tool by giving special attention to teaching activities at different universities and industrial cooperation, all of them within the industrial automation field. In an initial assessment, the use of an evaluation questionnaire tool within an academic platform allows members from different universities to collaborate and to share knowledge. By means of an example, the Barcelona Tech University in Spain provides support for teaching activities in automation to the ESPE University in Quito, Ecuador. A second case study example is presented in order to show how to detect sources of complexity in human supervisory control room tasks in wastewater treatment plants. The first results show that the interface is effective, easy to use and understand and can help engineering students in a training program to play the role of control room operators and the elicitation of sources of complexity can be useful to improve the interaction between humans and safety-critical systems in the industrial domain. 2012-10-08T11:17:51Z http://hdl.handle.net/2117/16581 Title: A field study with primary school children on perception of social presence and interactive behavior with a pet robot Authors: Heerink, Marcel; Díaz Boladeras, Marta; Albo-Canals, Jordi; Angulo Bahón, Cecilio; Barco, Alex; Casacuberta Bagó, Judit; Garriga Berga, Carles Abstract: This paper presents a study on measuring (1) how children experience a pet robot, (2) how they play with it and (3) how children’s perceptions on and interaction with pet robots are interrelated. The study features different types of subjective and objective techniques to assess the degree of perceived social entity from self-reports (i.e. questionnaires) and observed behavior. Three short questionnaires and an ad hoc code scheme of 15 low-level micro-behaviors were developed. 28 scholars aged 8 to 12 were observed at school during a play period with a Pleo robot and asked to answer the questionnaires. We found that the different questionnaire based methods were in line each other. Therefore, anyone of them can be used to measure the experience of a social entity. Play analyses showed that the two most prevalent behaviors were clearly social: petting the robot and showing it objects to engage in interaction. Moreover, children spent on average less than one per cent of the session time treating the robot as an artifact. However, significant covariation between the experience of a social entity and observed behavior could not be established. 2012-09-26T10:47:54Z