DIT - GAST - Capítulos de Monografías

Permanent URI for this collection

https://hdl.handle.net/10016/13135

Browse

Now showing 1 - 20 of 56

Analyzing Students' Persistence using an Event-Based Model
(CEUR-WS.org, 2019-06-27) Moreno Marcos, Pedro Manuel; Muñoz Merino, Pedro José; Alario Hoyos, Carlos; Delgado Kloos, Carlos; Agencia Estatal de Investigación (España); Comunidad de Madrid; Ministerio de Ciencia, Innovación y Universidades (España)
In education, persistence can be defined as the students' ability to keep on working on the assigned tasks (e.g., exercises) despite the difficulties. From previous studies, persistence might be an important factor in students' performance. However, these studies were limited because they only relied on students' self-reported data to measure persistence. This article aims to contribute with a novel model to measure persistence from students' logs, which is general enough to be applied to different educational platforms. In this work, persistence is measured taking students' interactions with automatic correction exercises. Simple metrics such as the average of students' attempts are not valid for a precise calculation of persistence since some exercises should count more for persistence as they have been done incorrectly many times but with some limit so that a single exercise cannot bias the indicator; or when a student answers correctly we should not add new attempts. In this paper, we propose a model to measure persistence on exercises which is valid to many digital online educational platforms. The analysis of students' persistence shows that there are not statistically significant differences of persistence between students who drop out the course or not, although persistence is shown to have a positive relationship with average grades in most of the cases. In contrast, persistence is not related to engagement with videos. These results provide an initial exploration about students' persistence, which can be important to understand how students behave and to properly adapt the course to students' needs.
An early warning dropout model in higher education degree programs: A case study in Ecuador
(CEUR-WS.org, 2020-09-14) Heredia-Jimenez, Vanessa; Jiménez Macías, Alberto Alejandro; Ortiz Rojas, Margarita; Imaz Marín, Jon; Moreno-Marcos, Pedro Manuel; Muñoz Merino, Pedro José; Delgado Kloos, Carlos; Ministerio de Economía, Comercio y Empresa; Comunidad de Madrid; European Commission
Worldwide, a significant concern of universities is to reduceacademic dropout rate. Several initiatives have been made to avoid thisproblem; however, it is essential to recognize at-risk students as soon aspossible. In this paper, we propose a new predictive model that can iden-tify the earliest moment of dropping out of a student of any semester inany undergraduate course. Unlike most available models, our solution isbased on academic information alone, and our evidence suggests that byignoring socio-demographics or pre-college entry information, we obtainmore reliable predictions, even when a student has only one academicsemester finished. Therefore, our prediction can be used as part of anacademic counseling tool providing the performance factors that couldinfluence a student to leave the institution. With this, the counselorscan identify those students and take better decisions to guide them andfinally, minimize the dropout in the institution. As a case study, we usedthe students¿ data of all undergraduate programs from 2000 until 2019from a public high education university in Ecuador.
An Initial Analysis of Prediction Techniques as a Support for the Flipped Classroom
(CEUR-WS.org, 2020-09-01) Rubio Fernández, Aarón; Moreno-Marcos, Pedro Manuel; Muñoz Merino, Pedro José; Delgado Kloos, Carlos; Comunidad de Madrid; Ministerio de Ciencia, Innovación y Universidades (España)
With the increasing use of active learning methodologies such as the Flipped Classroom (FC), many approaches have been taken to enhance the students' learning in such contexts. Prediction techniques can be used in combination with Learning Analytics (LA) dashboards for the improvement of the FC model. In this direction, we analyze some theoretical cases in which this approach can provide academic benefits (e.g. providing additional resources or re-designing the class). Furthermore, we present several initial ideas on how to combine two existing software tools, one which provides LA dashboards and the other that implements prediction techniques, that can be used successfully in such scenarios for the FC. This is a preliminary work for the joint use of prediction techniques and LA dashboards in FC contexts.
What Can You Do with Educational Technology that is Getting More Human?
(IEEE, 2019-04-09) Delgado Kloos, Carlos; Alario-Hoyos, Carlos; Muñoz Merino, Pedro José; Ibáñez Espiga, María Blanca; Estévez Ayres, Iria Manuela; Crespo García, Raquel; Comunidad de Madrid; Ministerio de Ciencia, Innovación y Universidades (España)
Technology is advancing at an ever-increasing speed. The backend capabilities and the frontend means of interaction are revolutionizing all kinds of applications. In this paper, we analyze how the technological breakthroughs seem to make educational interactions look smarter and more human. After defining Education 4.0 following the Industry 4.0 idea, we identify the key breakthroughs of the last decade in educational technology, basically revolving around the concept cloud computing, and imagine a new wave of educational technologies supported by machine learning that allows defining educational scenarios where computers interact and react more and more like humans.
Chrome Plug-in to Support SRL in MOOCs
(Springer, 2019-05-20) Alonso Mencía, María Elena; Alario-Hoyos, Carlos; Delgado Kloos, Carlos; Comunidad de Madrid; European Commission; Ministerio de Economía y Competitividad (España)
Massive Open Online Courses (MOOCs) have gained popularity over the last years, offering a learning environment with new opportunities and challenges. These courses attract a heterogeneous set of participants who, due to the impossibility of personal tutorship in MOOCs, are required to create their own learning path and manage one’s own learning to achieve their goals. In other words, they should be able to self-regulate their learning. Self-regulated learning (SRL) has been widely explored in settings such as face-to-face or blended learning environments. Nevertheless, research on SRL in MOOCs is still scarce, especially on supporting interventions. In this sense, this document presents MOOCnager, a Chrome plug-in to help learners improve their SRL skills. Specifically, this work focuses on 3 areas: goal setting, time management and selfevaluation. Each area is included in one of the 3 phases composing Zimmerman’s SRL Cyclical Model. In this way, the plug-in aims to support enrolees’ self-regulation throughout their complete learning process. Finally, MOOCnager was uploaded to the Chrome Web Store, in order to get a preliminary evaluation with real participants from 6 edX Java MOOCs designed by the Universidad Carlos III de Madrid (UC3M). Results were not conclusive as the use of the plug-in by the participants was very low. However, learners seem to prefer a seamless tool, integrated in the MOOC platform, which is able to assist them without any learner-tool interaction.
Predicting Learners' Success in a Self-paced MOOC Through Sequence Patterns of Self-regulated Learning
(Springer, 2018-09-03) Maldonado-Mahauad, Jorge; Perez Sanagustin, Maria Del Mar; Moreno-Marcos, Pedro Manuel; Alario-Hoyos, Carlos; Muñoz Merino, Pedro José; Delgado Kloos, Carlos; Ministerio de Economía y Competitividad (España); Ministerio de Educación, Cultura y Deporte (España)
In the past years, predictive models in Massive Open Online Courses (MOOCs) have focused on forecasting learners' success through their grades. The prediction of these grades is useful to identify problems that might lead to dropouts. However, most models in prior work predict categorical and continuous variables using low-level data. This paper contributes to extend current predictive models in the literature by considering coarse-grained variables related to Self-Regulated Learning (SRL). That is, using learners' self-reported SRL strategies and MOOC activity sequence patterns as predictors. Lineal and logistic regression modelling were used as a first approach of prediction with data collected from N = 2,035 learners who took a self-paced MOOC in Coursera. We identified two groups of learners: (1) Comprehensive, who follow the course path designed by the teacher; and (2) Targeting, who seek for the information required to pass assessments. For both type of learners, we found a group of variables as the most predictive: (1) the self-reported SRL strategies 'goal setting', 'strategic planning', 'elaboration' and 'help seeking'; (2) the activity sequences patterns 'only assessment', 'complete a video-lecture and try an assessment', 'explore the content' and 'try an assessment followed by a video-lecture'; and (3) learners' prior experience, together with the self-reported interest in course assessments, and the number of active days and time spent in the platform. These results show how to predict with more accuracy when students reach a certain status taking in to consideration not only low-level data, but complex data such as their SRL strategies.
Analysis of the Factors Influencing Learners' Performance Prediction With Learning Analytics
(2020-01-01) Moreno-Marcos, Pedro Manuel; Muñoz Merino, Pedro José; Delgado Kloos, Carlos; Comunidad de Madrid; Ministerio de Ciencia, Innovación y Universidades (España)
The advancement of learning analytics has enabled the development of predictive models to forecast learners' behaviors and outcomes (e.g., performance). However, many of these models are only applicable to specific learning environments and it is usually difficult to know which factors influence prediction results, including the predictor variables as well as the type of prediction outcome. Knowing these factors would be relevant to generalize to other contexts, compare approaches, improve the predictive models and enhance the possible interventions. In this direction, this work aims to analyze how several factors can make an influence on the prediction of students' performance. These factors include the effect of previous grades, forum variables, variables related to exercises, clickstream data, course duration, type of assignments, data collection procedure, question format in an exam, and the prediction outcome (considering intermediate assignment grades, including the final exam, and the final grade). Results show that variables related to exercises are the best predictors, unlike variables about forum, which are useless. Clickstream data can be acceptable predictors when exercises are not available, but they do not add prediction power if variables related to exercises are present. Predictive power was also better for concept-oriented assignments and best models usually contained only the last interactions. In addition, results showed that multiple-choice questions were easier to predict than coding questions, and the final exam grade (actual knowledge at a specific moment) was harder to predict than the final grade (average knowledge in the long term), based on different assignments during the course.
Assessing gait impairments based on auto-encoded patterns of mahalanobis distances from consecutive steps
(IOS Press, 2017) Muñoz Organero, Mario; Davies, Richard; Mawson, Sue
Insole pressure sensors capture the force distribution patterns during the stance phase while walking. By comparing patterns obtained from healthy individuals to patients suffering different medical conditions based on a given similarity measure, automatic impairment indexes can be computed in order to help in applications such as rehabilitation. This paper uses the data sensed from insole pressure sensors for a group of healthy controls to train an auto-encoder using patterns of stochastic distances in series of consecutive steps while walking at normal speeds. Two experiment groups are compared to the healthy control group: a group of patients suffering knee pain and a group of post-stroke survivors. The Mahalanobis distance is computed for every single step by each participant compared to the entire dataset sensed from healthy controls. The computed distances for consecutive steps are fed into the previously trained autoencoder and the average error is used to assess how close the walking segment is to the autogenerated model from healthy controls. The results show that automatic distortion indexes can be used to assess each participant as compared to normal patterns computed from healthy controls. The stochastic distances observed for the group of stroke survivors are bigger than those for the people with knee pain.
Proportional justified representation
(AAAI Press, 2017-02-13) Sánchez Fernández, Luis; Elkind, Edith; Lackner, Martin; Fernández García, Norberto; Arias Fisteus, Jesús; Basanta Val, Pablo; Skowron, Piotr
The goal of multi-winner elections is to choose a fixed-size committee based on voters’ preferences. An important concern in this setting is representation: large groups of voters with cohesive preferences should be adequately represented by the election winners. Recently, Aziz et al. proposed two axioms that aim to capture this idea: justified representation (JR) and its strengthening extended justified representation (EJR). In this paper, we extend the work of Aziz et al. in several directions. First, we answer an open question of Aziz et al., by showing that Reweighted Approval Voting satisfies JR for k = 3; 4; 5, but fails it for k >= 6. Second, we observe that EJR is incompatible with the Perfect Representation criterion, which is important for many applications of multi-winner voting, and propose a relaxation of EJR, which we call Proportional Justified Representation (PJR). PJR is more demanding than JR, but, unlike EJR, it is compatible with perfect representation, and a committee that provides PJR can be computed in polynomial time if the committee size divides the number of voters. Moreover, just like EJR, PJR can be used to characterize the classic PAV rule in the class of weighted PAV rules. On the other hand, we show that EJR provides stronger guarantees with respect to average voter satisfaction than PJR does.
A driving assistant to help people with mild cognitive ipairment or early stages of dementia
(International Society For Telemedicine & Ehealth (Isfteh), 2016) Muñoz Organero, Mario
Sudden or gradual loss of ability to drive a car - such as with sight loss and/or cognitive decline - can have significant impact on people's wellbeing and mental health. In the INLIFE project there is a desire to provide an integrated suite of assistive technologies to particularly support people with dementia or mild cognitive impairment. This paper presents a possible wearable technology to facilitate driving and monitor decline in driving capacity. The way we perform common daily activities is influenced by our cognitive abilities. Driving is one example of significant importance since the way we drive may have a potential impact in our safety. This paper proposes a framework to monitor and track the driving efficiency of a particular user in terms of energy consumption and perceived stress as a way to assess the user's cognitive abilities. A smartphone based personal recommender system is proposed to help the driver to drive better in high cognitive demanding situations as measured by the perceived stress by providing in advance information to help the driver better manage those situations. The perceived stress is estimated from the heart rate variability (HRV) signal from a wearable device. In the long term, the evolution in detected patterns in personal time series of perceived stress when driving in similar conditions could be used as an indicator of the severity and progress in mild cognitive impairment or early stages of dementia for particular patients.
Estimating the stress for drivers and passengers using deep learning
(CEUR-WS.org, 2016) Corcoba Magaña, Víctor; Muñoz Organero, Mario; Arias Fisteus, Jesús; Sánchez Fernández, Luis
The number of vehicles in circulation has become a problem both for safety and for the citizens health. Public transport is a solution to reduce its impact on the environment. One of the keys to encourag e users to use it is to improve comfort. On the other hand, numerous studies highlight that drivers are more likely to suffer physical and psychological illnesses due to the sedentary nature of this work and workload . In this paper, we propose a model to p redict the stress level on drivers and passengers. The solution is based on deep learning algorithms. The proposal employs the Heart Rate Variability (HRV) and telemetry from the vehicle in order to anticipate the incoming stress . It has been validated in a real environment on distinct routes. The results show that it predict s the stress by 86 % on drivers and 92% on passengers. This algorithm could be used to develop driving assistants that recommend actions to smooth driving, reducing the work load and the passenger stress.
SPOCs for Remedial Education: Experiences at the Universidad Carlos III de Madrid
(P.A.U. Education, 2014-02-10) Muñoz Merino, Pedro José; Méndez Rodríguez, Eva María; Delgado Kloos, Carlos
The Universidad Carlos III de Madrid has been offering several face-to-face remedial courses for freshmen to review or learn concepts and practical skills that they should know before starting their degree programme. During the last two years, our University has adopted MOOC-like technologies to support some of these courses so that a "fipping the classroom" methodology can be applied to a particular small educational context. This paper gathers a list of issues and challenges encountered when using Khan Academy technologies for small private online courses (SPOCs). These issues and challenges include the absence of a single platform that supports all the requirements, the need for integration of different learning platforms, the complexity of the authoring process, the need for an adaptation of gamifcation during the learning process and the adjustment of the learning analytics functionality. In addition, some lessons learned are presented, as well as specifc actions taken in response, where MOOCs do not replace teachers and classrooms for these remedial courses, but improve their effectiveness.
Eco-driving: energy saving based on driver behavior
(Juan Antonio Ortega Ramírez, Alejandro Fernández-Montes, Juan Antonio Álvarez, 2015) Corcoba Magaña, Víctor; Muñoz Organero, Mario
The number of vehicles has grown in recent years. As a result, it has increased the fuel consumption and the emission of gaseous pollutants. The emission of gaseous pollutants causes more deaths than traffic accidents. On the other hand, the energy resources are limited and the increase in demand causes them even more expensive. In addition, the percentage of old vehicles is very high. Eco-driving is a good solution in order to minimize the fuel consumption because it is independent of the vehicle age. In this paper, a driving assistant is presented. This solution allows the user acquires knowledge about eco-driving. Unlike other solutions, our proposal adapts the recommendations to the user profile. It also provides information in advance such as: optimal average speed, anomalous events, deceleration pattern, and so on. These recommendations prevent that the user performs inefficient actions. In these type of systems, motivation is very important. Drivers lose the interest over time. To solve this problem, we employ gamification techniques that contribute to avoid drivers coming back to their previous driving habits
The impact of using gamification on the eco-driving learning
(Springer International Publishing, 2014-06) Corcoba Magaña, Víctor; Muñoz Organero, Mario
This paper analyses and validates the impact of using gamification techniques for improving eco-driving learning. The proposal uses game mechanisms such as the score and achievements systems in order to encourage the driver to drive efficiently. The score is calculated using fuzzy logic techniques that allow us to evaluate the driver in a similar way as a human being would do. We also define the eco-driving tips that are issued while driving in order to help the driver to improve the fuel consumption. Every time the system detects an inefficient action of the driver to a previously known situation such as a bad reaction to a detected traffic sign or a detected traffic accident, it warns the user. The proposal is validated using 14 different drivers performing more than 300 drives with 5 different models of vehicles on 4 different regions of Spain. The conclusions show a positive correlation in the use of gamification techniques and the application of the proposed of eco-driving tips, especially for aggressive drivers. Furthermore, these techniques contribute to avoid drivers coming back to their previous driving habits.
Reducing Stress and Fuel Consumption Providing Road Information
(Springer International Publishing Switzerland, 2015-05) Corcoba Magaña, Víctor; Muñoz Organero, Mario
In this paper, we propose a solution to reduce the stress level of the driver, minimize fuel consumption and improve safety. The system analyzes the driving and driver workload during the trip. If it discovers an area where the stress increases and the driving style is worse from the point of view of energy efficiency, a photo is taken and is saved along with its location in a shared database. On the other hand, the solution warns the user when is approaching a region where the driving is difficult (high fuel consumption and stress) using the shared database. In this case, the proposal shows on the screen of the mobile device the image captured previously of the area. The aim is that driver knows in advance the driving environment. Therefore, he or she may adjust the vehicle speed and the driver workload decreases. Data Envelopment Analysis is used to estimate the efficiency of driving and driver workload in each area. We employ this method because there is no preconceived form on the data in order to calculate the efficiency and stress level. A validation experiment has been conducted with 6 participants who made 96 driving tests in Spain. The system reduces the slowdowns (38 %), heart rate (4.70 %), and fuel consumption (12.41 %). The proposed solution is implemented on Android mobile devices and does not require the installation of infrastructure on the road. It can be installed on any model of vehicle.
Computer Assisted Assessment within 3D Virtual Worlds
(Universidad Carlos III de Madrid, 2011) Ibáñez Espiga, María Blanca; Morillo, Diego; Santos, Patricia; Perez Calle, David; García Rueda, José Jesús; Hernández-Leo, Davinia; Delgado Kloos, Carlos
3D Virtual Worlds are currently been explored as learning environments due to their capabilities to promote learner motivation. Most of the research has been focused on the deployment of learning strategies on them. However, a crucial component of the teaching-learning process: the assessment has been neglected. In this work, we present an architecture that integrates an engine QTI-compliant with a 3D virtual world platform. The rich set of interactions that can occur in a 3D virtual environment is mapped onto the 2D format of the QTI-specification. As result, our solution allows a computer assisted assessment into a 3D virtual world using metaphors adapted to this rich interface. We present an assessment experience deployed on our architecture. The results of the evaluation show satisfaction of learners in using test assessment questions within 3DVWs and improvement in learning outcomes.
A Visual Specification Language for Model-to-Model Transformations
(Ieee - The Institute Of Electrical And Electronics Engineers, Inc, 2010-09) Guerra, Esther; Lara, Juan de; Kolovos, Dimitris; Paige, Richard F.
While interaction patterns are becoming widespread in the field of interface design, their definitions do not enjoy a common standard yet, as is for software patterns. Moreover, patterns are developed for diverse design aspects, reflecting the complexity of the field. As a consequence, research on formalization of interaction patterns is not developed, and few attempts have been made to extend techniques developed for design pattern formalization. We show here how an extension to our recent approach to pattern formalization can be usefully employed to formalize some classes of interaction patterns, to express relations among them, and to detect conflicts.
Lightweight Executability Analysis of Graph Transformation Rules
(Ieee - The Institute Of Electrical And Electronics Engineers, Inc, 2010) Planas, Elena; Cabot, Jordi; Gómez, Cristina; Guerra, Esther; Lara, Juan de
Domain Specific Visual Languages (DSVLs) play a cornerstone role in Model-Driven Engineering (MDE), where (domain specific) models are used to automate the production of the final application. Graph Transformation is a formal, visual, rule-based technique, which is increasingly used in MDE to express in-place model transformations like refactorings, animations and simulations. However, there is currently a lack of methods able to perform static analysis of rules, taking into account the DSVL meta-model integrity constraints. In this paper we propose a lightweight, efficient technique that performs static analysis of the weak executability of rules. The method determines if there is some scenario in which the rule can be safely applied, without breaking the meta-model constraints. If no such scenario exists, the method returns meaningful feedback that helps repairing the detected inconsistencies.
Towards Parallel Educational Worlds
(Ieee - The Institute Of Electrical And Electronics Engineers, Inc, 2011) Delgado Kloos, Carlos; Fernández Panadero, María Carmen; Ibáñez Espiga, María Blanca; Muñoz Organero, Mario; Pardo Sánchez, Abelardo
Augmented Reality, 3D virtual worlds, etc.: the technology has evolved tremendously and so has its application to the field of education. Digital technologies have advanced to the point, where we are reproducing digitally more and more aspects of our life. We have parallel worlds: on the one hand the real world, and on the other virtual worlds, that can in fact be linked to the real one. They have different properties, but they can enrich and complement each other. In this paper, we explore the possibilities and challenges of these parallel worlds for educational uses.
Multi-User 3D Virtual Environment for Spanish Learning: A Wonderland Experience
(Ieee - The Institute Of Electrical And Electronics Engineers, Inc, 2010) Ibáñez Espiga, María Blanca; García Rueda, José Jesús; Galán, Sergio; Maroto, David; Morillo, Diego; Delgado Kloos, Carlos
In this paper, we describe a 3D virtual collaborative system designed for the learning of Spanish as a second language. Several initiatives for second language learning in 3D virtual worlds exploiting immersive, interactive and motivating features of these worlds have been carried out successfully during the last years. However, these systems tend to be sometimes too rigid from a pedagogical point of view, requiring the presence of a teacher. We have used the Wonderland development toolkit to deploy a 3D virtual learning environment, which is flexible enough to allow learners to improve their language skills with minimum teacher's help, setting up an instructional sequence in which fostered, motivating, and pre-designed collaboration is the key for self-learning. The environment includes technical issues such as natural text chatting with synthetic characters, textual tagging of virtual objects, automatic reading of texts, and the integration of a 3D mouse in learning sequences in order to exploit the capabilities of 3D virtual worlds.

Browse

Recent Submissions