Papers by Raquel Torres Peralta
This research work seeks to improve the output of a data mining algorithm that supports diabetic ... more This research work seeks to improve the output of a data mining algorithm that supports diabetic patients' care. The model that is currently in operation uses three variables that are obtained from a general medical appointment database. This work aims to find other characteristics in the database to add them to those already considered to better describe patients to provide more accurate information. The article shows the process followed to improve the results of a k-means grouping algorithm for the follow-up process of diabetic patients. We present the process of defining the considered characteristics that were not part of the model, to analyze and eventually add them. A qualitative comparison between the algorithms is shown and the findings are explained during the analysis of the studied variables, in relation to sex and age of the patients.
Communications in Computer and Information Science, 2018
This article presents a method that extracts relevant concepts automatically, consisting of one o... more This article presents a method that extracts relevant concepts automatically, consisting of one or several words, whose main contribution is that it does so from a single document of any domain, regardless of its length; however, documents of short length are used (which are the most frequent to obtain on the web) to perform the work. This research was conducted for documents written in Spanish and was tested in multiple randomized domains to compare their results. For this, an algorithm was used to automatically identify syntactic patterns in the document. This work uses the previous work of [1] to obtain its results. This algorithm is based on statistical approximations and on the length of the identifiable patterns contained in the document, applies certain heuristic that can enhance or decrease the patterns' choice according to the selection of one of the 5 methods that are processed (M1 to M5), with these patterns the candidate concepts are obtained, which go through another evaluation process that will obtain the final concepts. This proposal presents at least four advantages: (1) It is multi-domain, (2) It is independent of the text length, (3) It can work with one or more documents and (4) It allows the discarding of garbage or undesirable patterns from the beginning. The method was implemented in 11 different domains and its results range varies between 58%-70% of precision and 25%-46% of recall.
The answering of any test represents a challenge for students; however, foreign students whose fi... more The answering of any test represents a challenge for students; however, foreign students whose first language is not English have to deal with the difficulty of the understanding of a series of questions written on a different language in addition of the effort required to solve the problem. In this study, we recorded the behavior of the brain signals of 16 students, 10 whom first language was English and 6 who were English learners, and used two supervised classification algorithms in order to identify the students’ language proficiency. The results shown that in both approaches, harder problems which required longer time to be responded had a higher accuracy rate; however, more tests are needed in order to understand the physical processing of written math text problem and the difference among both groups.
EPISTEMUS, 2017
En este documento se realiza un análisis que permita identificar las principales causas de desgas... more En este documento se realiza un análisis que permita identificar las principales causas de desgaste asociados con los neumáticos de los tractocamiones de carga. Se realiza una evaluación sobre el desempeño y rentabilidad de neumáticos nuevos y reparados total o parcialmente. La obtención de estadísticas es una valiosa herramienta que apoya la toma de decisiones,permite realizar ajustes en los procesos para lograr incrementos en la productividad y rentabilidad de la empresa. Además, los beneficios de un mejor cuidado de los neumáticos no se limitan a lo económico, sino también a un impacto ambiental al reducir los desechos.
Proceedings of the 2016 International Conference on Quantitative InfraRed Thermography, 2016
The motivation of the present work is to develop a simple grading scale (mild, moderate or severe... more The motivation of the present work is to develop a simple grading scale (mild, moderate or severe) for estimation of an asymmetry level of breast thermograms. The scale should help to distribute test subjects into three groups. The first group, with mild asymmetry, will be supposed to have no thermographic abnormalities. Persons from the second group, moderate asymmetry, will be asked to receive a second thermography exam within 3 to 6 months. The third group, individuals with severe asymmetry, will be referred to oncology doctor and receive the second thermography exam within 2 to 3 months.
Proceedings of the 2015 Asia International Conference on Quantitative InfraRed Thermography, 2015
Proceedings of the 2015 Asia International Conference on Quantitative InfraRed Thermography, 2015
Proceedings of the 2015 Asia International Conference on Quantitative InfraRed Thermography, 2015
Lecture Notes in Computer Science, 2013
Multi-touch tablets can offer a large, collaborative space where several users can work on a task... more Multi-touch tablets can offer a large, collaborative space where several users can work on a task at the same time. However, the lack of privacy in these situations makes standard password-based authentication easily compromised. This work presents a new gesture-based authentication system based on users' unique signature of touch motion when drawing a combination of onestroke gestures following two different policies, one fixed for all users and the other selected by a model of control to maximize the expected long-term information gain. The system is able to achieve high user recognition accuracy with relatively few gestures, demonstrating that human touch patterns have a distinctive "signature" that can be used as a powerful biometric measure for user recognition and personalization.
Multi-touch tablets allow users to interact with computers through intuitive, natural gestures an... more Multi-touch tablets allow users to interact with computers through intuitive, natural gestures and direct manipulation of digital objects. One advantage of these devices is that they can offer a large, collaborative space where several users can work on a task at the same time. However the lack of privacy in these situations makes standard password-based authentication easily compromised. We therefore propose a new gesture-based authentication system based on each users' unique signature of touch motion. Our technique has two key features. First, at each step in authentication the system prompts the user to make a specific gesture selected to maximize the expected long-term information gain. Hence the order of gestures can be different every time, and does not require the user to memorize a specific pattern. Second, each gesture is integrated using a novel hierarchical probabilistic model, allowing the system to accept or reject a user after a variable number of gestures, often in as few as five gestures. This touch-based approach allows the user to accurately authenticate without the need to cover their hand or look over their shoulder. It also allows the system to adapt to the diversity in touch styles for any particular user population. We tested this method using a set of samples were collected under realworld conditions in a business office, using a touch tablet that was used on a near daily basis by users familiar with the device. Despite the lack of sophisticated, high-precision equipment or noise suppression techniques, our system is able to achieve extremely high user recognition accuracy with relatively few gestures, demonstrating that human touch patterns have a distinctive "signature" can be used as a powerful biometric measure for user recognition and personalization.
international conference on user modeling adaptation and personalization, Jul 11, 2011
Our goal is to develop methods for non-experts to teach complex behaviors to autonomous agents (s... more Our goal is to develop methods for non-experts to teach complex behaviors to autonomous agents (such as robots) by accommodating "natural" forms of human teaching. We built a prototype interface allowing humans to teach a simulated robot a complex task using several techniques and report the results of 44 human participants using this interface. We found that teaching styles varied considerably but can be roughly categorized based on the types of interaction, patterns of testing, and general organization of the lessons given by the teacher. Our study contributes to a better understanding of human teaching patterns and makes specific recommendations for future human-robot interaction systems.
Our goal is to develop methods for non-experts to teach complex
behaviors to autonomous agents (s... more Our goal is to develop methods for non-experts to teach complex
behaviors to autonomous agents (such as robots) by accommodating “natural” forms of human teaching.We built a prototype interface allowing humans to teach a simulated robot a complex task using several techniques and report the results of 44 human participants using this interface. We found that teaching styles varied considerably but can be roughly categorized based on the types of interaction, patterns of testing, and general organization of the lessons given by the teacher. Our study contributes to a better understanding of human teaching patterns and makes specific recommendations for future human-robot interaction systems.
cs.arizona.edu
Currently, most systems for human-robot teaching allow only one mode of teacher-student interacti... more Currently, most systems for human-robot teaching allow only one mode of teacher-student interaction (e.g., teaching by demonstration or feedback), and teaching episodes have to be carefully set-up by an expert. To understand how we might integrate multiple, interleaved forms of human instruction into a robot learner, we performed a behavioral study in which 44 untrained humans were allowed to freely mix interaction modes to teach a simulated robot (secretly controlled by a human) a complex task. Analysis of transcripts showed that human teachers often give instructions that are nontrivial to interpret and not easily translated into a form useable by machine learning algorithms. In particular, humans often use implicit instructions, fail to clearly indicate the boundaries of procedures, and tightly interleave testing, feedback, and new instruction. In this paper, we detail these teaching patterns and discuss the challenges they pose to automatic teaching interpretation as well as the machine-learning algorithms that must ultimately process these instructions. We highlight the challenges by demonstrating the difficulties of an initial automatic teacher interpretation system.
2011 AAAI Spring …, Jan 1, 2011
Humans naturally use multiple modes of instruction while teaching one another. We would like our ... more Humans naturally use multiple modes of instruction while teaching one another. We would like our robots and artificial agents to be instructed in the same way, rather than programmed. In this paper, we review prior work on human instruction of autonomous agents and present the results of three studies investigating how multiple instruction modes are used by humans. We describe our Bootstrapped Learning User Interface, a prototype multi-instruction interface informed by our human-user studies.
Uploads
Papers by Raquel Torres Peralta
behaviors to autonomous agents (such as robots) by accommodating “natural” forms of human teaching.We built a prototype interface allowing humans to teach a simulated robot a complex task using several techniques and report the results of 44 human participants using this interface. We found that teaching styles varied considerably but can be roughly categorized based on the types of interaction, patterns of testing, and general organization of the lessons given by the teacher. Our study contributes to a better understanding of human teaching patterns and makes specific recommendations for future human-robot interaction systems.
behaviors to autonomous agents (such as robots) by accommodating “natural” forms of human teaching.We built a prototype interface allowing humans to teach a simulated robot a complex task using several techniques and report the results of 44 human participants using this interface. We found that teaching styles varied considerably but can be roughly categorized based on the types of interaction, patterns of testing, and general organization of the lessons given by the teacher. Our study contributes to a better understanding of human teaching patterns and makes specific recommendations for future human-robot interaction systems.