User-adaptive Systems Research Articles

Improving Teacher Training Through Emotion Recognition and Data Fusion

ABSTRACTThe quality of education hinges on the proficiency and training of educators. Due to the importance of teacher training, the innovative platform Teacher Moments creates simulated classroom scenarios. In this scenario‐based learning, confusion is an important indicator to detect users who struggle with the simulations. Through Teacher Moments, we gathered 7975 audio recording responses from participants who self‐labelled their recordings according to whether they sounded confused. Our dataset stands out for its size, for not including actor‐generated audio, and for measuring confusion, a neglected emotion in artificial intelligence (AI). Our experiments tested unimodal approaches and feature‐level, model‐level and decision‐level fusion. Feature‐level fusion demonstrated superior performance to unimodal methods, achieving a balanced accuracy of 0.6607 on the test set. This outcome highlights the necessity for further investigation in the overlooked area of confusion detection, particularly employing realistic datasets like the one used in this study and exploring new methods. Beyond teacher training, the insights of this research also extend to other directions, such as other professionals making critical decisions, user interface design or adaptive learning systems.

MENCIPTAKAN KOTA CERDAS: MEMANFAATKAN AI (ARTIFICIAL INTELLIGENCE) UNTUK PENGEMBANGAN RUANG PUBLIK PERKOTAAN YANG CERDAS

As cities continue to grow and face increasing challenges, the concept of smart cities has emerged as a solution to create sustainable, efficient, and livable urban environments. One crucial aspect of smart cities is the development of intelligent urban public spaces that cater to the diverse needs and preferences of residents and visitors. This paper explores the role of AI in the development of such spaces, focusing on how AI technologies can enhance their design, functionality, and user experience. It examines the application of AI in areas such as data analytics, user behavior analysis, sensor networks, and adaptive systems to create dynamic and responsive public spaces. Furthermore, it discusses the benefits, challenges, and potential ethical considerations associated with harnessing AI for intelligent urban public spaces development. Through this exploration, the paper highlights the transformative potential of AI in shaping the future of urban environments and fostering inclusive and sustainable communities.

Telepresence Social Robotics towards Co-Presence: A Review

Telepresence robots are becoming popular in social interactions involving health care, elderly assistance, guidance, or office meetings. There are two types of human psychological experiences to consider in robot-mediated interactions: (1) telepresence, in which a user develops a sense of being present near the remote interlocutor, and (2) co-presence, in which a user perceives the other person as being present locally with him or her. This work presents a literature review on developments supporting robotic social interactions, contributing to improving the sense of presence and co-presence via robot mediation. This survey aims to define social presence, co-presence, identify autonomous “user-adaptive systems” for social robots, and propose a taxonomy for “co-presence” mechanisms. It presents an overview of social robotics systems, applications areas, and technical methods and provides directions for telepresence and co-presence robot design given the actual and future challenges. Finally, we suggest evaluation guidelines for these systems, having as reference face-to-face interaction.

Measuring and modeling context-dependent preferences for hearing aid settings

Despite having individual perceptual preferences toward sounds, hearing aid users often end up with default hearing aid settings that have no contextual awareness. However, the introduction of smartphone-connected hearing aids has enabled a rethinking of hearing aids as user-adaptive systems considering both individual and contextual differences. In this study, we aimed to investigate the feasibility of such context-aware system for providing hearing aid users with a number of relevant hearing aid settings to choose from. During normal real-world hearing aid usage, we applied a smartphone-based method for capturing participants’ listening experience and audiological preference for different intervention levels of three audiological parameters (Noise Reduction, Brightness, Soft Gain). Concurrently, we collected contextual data as both self-reports (listening environment and listening intention) and continuous data logging of the acoustic environment (sound pressure level, signal-to-noise ratio). First, we found that having access to different intervention levels of the Brightness and Soft Gain parameters affected listening satisfaction. Second, for all three audiological parameters, the perceived usefulness of having access to different intervention levels was significantly modulated by context. Third, contextual data improved the prediction of both explicit and implicit intervention level preferences. Our findings highlight that context has a significant impact on hearing aid preferences across participants and that contextual data logging can help reduce the space of potential interventions in a user-adaptive system so that the most useful and preferred settings can be offered. Moreover, the proposed mixed-effects model is suitable for capturing predictions on an individual level and could also be expanded to predictions on a group level by including relevant user features.

Medication recommendation system for online pharmacy using an adaptive user interface

BackgroundThis article proposes a prototype of a user-adaptive system for helping patients to obtain their ambulatory prescribed medications when purchasing online in a more convenient manner than traditional methods, and the adoption of artificial intelligence to achieve improvements. The system developed simulates an online pharmacy with an introductory adaptive user interface using Bayesian user modeling for predicting the medication needs of patients. This program is used to show its step-by-step design and functioning. MethodsThe introductory adaptive user interface was developed on Visual C++ of Microsoft Visual Studio. The patient model acquisition and application implementing the learning and inference was performed with a Bayesian Network. The Bayesian network was elaborated with the GeNIe Modeler software, Version 2.3.R4, provided by BayesFusion, LLC. Synthetic data from a synthetically generated dataset of anonymous patients was used. The performance of the system was evaluated through simulations using testing data from the synthetic dataset. The Accuracy of predictions was analyzed. ResultsThe Average accuracy was estimated with the average correct recommendations of medications, for different numbers of purchased medications per session. The Average accuracy increased with the number of purchased medications, from 86.3529% up to 92.5303%. The Average wrong recommendations decreased with the increase in the number of purchased medications, from an average of 3.4117 up to 1.5686. ConclusionThe system quickly and consistently attained high accuracy in predicting the medication categories needed by the patients, potentially being able to save time and effort for the patients by relying on the system's recommendations.

Predicting Visual Search Task Success from Eye Gaze Data as a Basis for User-Adaptive Information Visualization Systems

Information visualizations are an efficient means to support the users in understanding large amounts of complex, interconnected data; user comprehension, however, depends on individual factors such as their cognitive abilities. The research literature provides evidence that user-adaptive information visualizations positively impact the users’ performance in visualization tasks. This study attempts to contribute toward the development of a computational model to predict the users’ success in visual search tasks from eye gaze data and thereby drive such user-adaptive systems. State-of-the-art deep learning models for time series classification have been trained on sequential eye gaze data obtained from 40 study participants’ interaction with a circular and an organizational graph. The results suggest that such models yield higher accuracy than a baseline classifier and previously used models for this purpose. In particular, a Multivariate Long Short Term Memory Fully Convolutional Network shows encouraging performance for its use in online user-adaptive systems. Given this finding, such a computational model can infer the users’ need for support during interaction with a graph and trigger appropriate interventions in user-adaptive information visualization systems. This facilitates the design of such systems since further interaction data like mouse clicks is not required.

Accelerometer-based human activity recognition using 1D convolutional neural network

Human activity recognition (HAR) is an important research field with a variety of applications in healthcare monitoring, fitness tracking and in user-adaptive systems in smart environments. The performance of the activity recognition system is highly dependent on the features extracted from the sensor data which makes the selection of appropriate features a very important part of HAR. A 1D CNN model trained on accelerometer data is suggested in the paper for automatic feature extraction in a HAR system. A semi-automatic approach is used that effectively and efficiently determines the number of convolutional layers in the network, the number of kernels and the size of the kernels. The experimental results show that the suggested model outperforms several existing recognition approaches that use the same data set.

Visuo-motor adaptation during interaction with a user-adaptive system

Visuo-motor adaptation during interaction with a user-adaptive system

BUM: Bayesian User Model for Distributed Learning of User Characteristics From Heterogeneous Information

This paper presents a Bayesian user model able to learn and estimate user characteristics in a distributed manner using heterogeneous information. A unified user representation is obtained from an inference process, receiving a set of independently estimated user characteristics from different sources. The independence of characteristic models enables the system to be modular, with each module estimating one characteristic. The proposed model is iterative, fusing new observations, and measurements with previous information in a process regulated entropy. The system allows diverse implementations, such as the combination of multiple robots with a cloud infrastructure or distributed ambient sensors. This paper aims to enable the system to perform online learning while interacting with users. The system is also able to obtain a correct user representation from heterogeneous information, even when some user characteristics cannot be computed. To demonstrate its functionality, the system is tested on two experimental datasets, obtained from simulated experiments and with real users. This technique advances the state of the art in the areas of AAL and user-adaptive systems, and in cloud-connected robots and Internet of Things, allowing for these heterogeneous and naturally distributed teams of devices to better model their users, potentially achieving higher interaction autonomy.

User-Adaptive Interaction in Social Robots: A Survey Focusing on Non-physical Interaction

This work presents a survey on the usage of user-adaptive techniques for human interface with Social Robots, with focus on non-physical interaction. The work is based on an analysis of a number of recent scientific works, and aims to uncover existing scientific and technological gaps which can serve as basis for future research and development work. User-adaptive systems consist of autonomous agents that are able to use some manner of information on their user in order to adapt to them. Through their adaptive nature, these systems have been shown to be easier to accept by end-users, and to lead to improvements in a myriad of objective and subjective performance measurements. Thus, in the context of a growing domestic Social Robot industry, it becomes of key importance to study the scientific and technological frontiers of this field. In order to uncover potential lines of future research, we propose a taxonomy for the classification of works, which we use to analyse the works under survey, exposing the current scientific frontiers of the area. Aiming to establish the overall readiness of the field, we also analyse the maturity of the works under survey, exposing the current technological level of the techniques at hand and discussing a number of technological challenges.

Real-time human activity recognition from accelerometer data using Convolutional Neural Networks

With a widespread of various sensors embedded in mobile devices, the analysis of human daily activities becomes more common and straightforward. This task now arises in a range of applications such as healthcare monitoring, fitness tracking or user-adaptive systems, where a general model capable of instantaneous activity recognition of an arbitrary user is needed. In this paper, we present a user-independent deep learning-based approach for online human activity classification. We propose using Convolutional Neural Networks for local feature extraction together with simple statistical features that preserve information about the global form of time series. Furthermore, we investigate the impact of time series length on the recognition accuracy and limit it up to 1s that makes possible continuous real-time activity classification. The accuracy of the proposed approach is evaluated on two commonly used WISDM and UCI datasets that contain labeled accelerometer data from 36 and 30 users respectively, and in cross-dataset experiment. The results show that the proposed model demonstrates state-of-the-art performance while requiring low computational cost and no manual feature engineering.

Interview with Paul De Bra

Prof. dr. Paul De Bra is Full Professor in the Department of Computer Science at the Eindhoven University of Technology (TU/e). He holds a doctorate in mathematics and computer science from the University of Antwerp, Belgium (1987), with a thesis on Horizontal Decomposition in the Relational Database Model. During 1988 and 1989 he was a post-doctoral researcher at AT&T Bell Laboratories in Murray Hill, New Jersey, studying principles and technology for WYSIYWYG interfaces for document processing. Since December 1989 he has been working at the TU/e, first as an associate professor, and since August 1996 as a full professor and chair of the Information Systems group. Since he started working at the TU/e the research has shifted to the area of hypertext and hypermedia, with research in adaptive Web-based systems. He initiated and led the AHAM (reference model), AHA! and GRAPPLE projects on adaptive Technology-Enhanced Learning, the NWO CHIP project on adaptation and personalization in Cultural Heritage, and numerous other activities to stimulate research and development of adaptive hypermedia technology and applications. He is scientific director of the Dutch Research School for Information and Knowledge Systems and is president of User Modeling Inc. (the non-profit organization for User Modeling and Adaptive Systems). In "real" life Paul De Bra plays, repairs, composes and arranges for the accordion (and takes pictures at conferences too). "Paul De Bra" and "accordion" currently yields 76 hits on YouTube whereas "Paul De Bra" and "hypertext" yields 0. (On Google the hypertext query is more successful.)

Evaluating intelligent knowledge systems: experiences with a user-adaptive assistant agent

This article examines experiences in evaluating a user-adaptive personal assistant agent designed to assist a busy knowledge worker in time management. We examine the managerial and technical challenges of designing adequate evaluation and the tension of collecting adequate data without a fully functional, deployed system. The CALO project was a seminal multi-institution effort to develop a personalized cognitive assistant. It included a significant attempt to rigorously quantify learning capability, which this article discusses for the first time, and ultimately the project led to multiple spin-outs including Siri. Retrospection on negative and positive experiences over the 6 years of the project underscores best practice in evaluating user-adaptive systems. Lessons for knowledge system evaluation include: the interests of multiple stakeholders, early consideration of evaluation and deployment, layered evaluation at system and component levels, characteristics of technology and domains that determine the appropriateness of controlled evaluations, implications of ‘in-the-wild’ versus variations of ‘in-the-lab’ evaluation, and the effect of technology-enabled functionality and its impact upon existing tools and work practices. In the conclusion, we discuss—through the lessons illustrated from this case study of intelligent knowledge system evaluation—how development and infusion of innovative technology must be supported by adequate evaluation of its efficacy.

Dynamic Thermal Model for Hybrid Photovoltaic Panels

Photovoltaic technology is a consolidated solution for electricity production in residential, commercial and industrial facilities. Despite that, the solar-to-electrical energy conversion is still low and many efforts are still needed in order to increase the panel efficiency. Since the photovoltaic modules performance decreases when the cell temperature increases, solutions have been investigated to cool down the panel with a refrigerated thermal plate installed on the PV rear surface. In standard applications, the photovoltaic thermal panel (PV/T) systems can produce both thermal and electrical energy and they are typically used for domestic hot water (DHW) production. Moreover, hybrid solar panels must be coupled with a heat pump system at the evaporator side, realizing a photovoltaic solar assisted heat pump (PV-SAHP) plant. Generally, a careful design of these integrated systems must involve accurate control strategy to optimize energy savings during operation. In fact, the behavior of the PV/T panels is heavily influenced by the very quick variation of the external conditions during daytime (mainly due to solar irradiation), affecting the working conditions of the other equipment. Therefore, models able to reproduce the dynamic behavior of the PV/T panels represent a flexible tool for developing innovative and user-adapted system control criteria, in a global system optimization perspective. In this context, the aim of the present paper is to describe a simplified numerical model able to reproduce the short time dynamic behavior of the PV/T panel. The model has been validated using experimental data which have been collected during outdoor tests conducted at the University of Genoa using a prototype realized by retrofitting a commercial PV collector. Several simulations have been performed by comparing the PV/T outlet temperature provided by the numerical model against experimental data.

Methods and Techniques of Adaptive Web Accessibility for the Blind and Visually Impaired

Massive volumes of information shared on the web together with linked open data initiatives and exabytes of data generated through social networks frequently disorientates and confuses web users in their everyday interaction. Additionally, web users constitute a highly heterogeneous entities with different needs and requirements. Considering this, user adaptive software systems have been developed as a new application approach to ease the interaction between users and web information with the intent to bridge the gap between such presentation and navigational pitfalls. The user adaptive interaction is especially useful when considering a marginalized group such as blind and visually impaired users. This paper attempts at providing an overview of a state-of-the-art survey concerning adaptive interaction between users and web information space with special emphasis on exploiting methods and techniques for adaptive web accessibility for blind and visually impaired people. Likewise, the exploration of possibilities of new methods and techniques for user adapted interaction for blind and visually impaired is highly required in order to alleviate the accessibility according to Web Content Accessibility Guidelines (WCAG). Recent studies indicate that even when the guidelines are implemented on web information systems, there is little indication that people with disabilities will gain better accessibility. To address these issues, we introduce various visual and auditory approaches to extend such adaptive methods and techniques for blind and visually impaired.

Fuzzy Logic Based Multi User Adaptive Test System

The present proliferation of e-learning has been actively underway for the last 10 years. Current research in Adaptive Testing System focuses on the development of psychometric models with items selection strategies applicable to adaptive testing processes. The key aspect of proposed Adaptive Testing System is to develop an increasingly sophisticated latent trait model which can assist users in developing and enhancing their skills. Computerized Adaptive Test (CAT) System requires a lot of investment in time and effort to develop analyze and administrate an adaptive test. In this paper a fuzzy logic based Multi User Adaptive Test System (MUATS) is developed. Which is a Short Messaging Service (SMS) based System, currently integrated with GSM network based on the new psychometric model in education assessment. MUATS is not only a platform independent Adaptive Test System but also it eases the evaluation effort for adaptive test process. It further uses fuzzy logic to pick the most appropriate question from the pool of database for a specific user to be asked which makes the overall system an intelligent one.

Biomedical user adaptive system for smart environments

Ubiquitous computing (UbiCom) represents an umbrella for Smart Environments, Smart Devices and Smart Interactions. UbiCom also covers an area of user adaptive systems (UAS), which are growing in past time along with expansion of intelligent ubiquitous systems to every piece of embedded device around us. These systems provide many of interesting and useful services to support our day life needs in all areas include biomedical world of our bodies. Every such service needs however to provide their results by some kind of interface. Therefore the need of a simple, intuitive and graphically attractive interface is much appreciated. This paper is focused to several areas of user interface design, user interface adaptivity and visualisation. In all areas of developed system is implemented secured ways.

User model interoperability: a survey

Nowadays a large number of user-adaptive systems has been developed. Commonly, the effort to build user models is repeated across applications and domains, due to the lack of interoperability and synchronization among user-adaptive systems. There is a strong need for the next generation of user models to be interoperable, i.e. to be able to exchange user model portions and to use the information that has been exchanged to enrich the user experience. This paper presents an overview of the well-established literature dealing with user model interoperability, discussing the most representative work which has provided valuable solutions to face interoperability issues. Based on a detailed decomposition and a deep analysis of the selected work, we have isolated a set of dimensions characterizing the user model interoperability process along which the work has been classified. Starting from this analysis, the paper presents some open issues and possible future deployments in the area.

The evaluation of adaptive and personalised information retrieval systems: a review

A current problem with the research of adaptive systems is the inconsistency of evaluation applied to the adaptive systems. However, evaluating an adaptive system is a difficult task due to the complexity of such systems. Evaluators need to ensure correct evaluation methods and measurement metrics are used. This paper reviews a variety of evaluation techniques applied in adaptive and user-adaptive systems. More specifically, it focuses on the user-centred evaluation of adaptive systems such as personalised recommender systems and adaptive information retrieval systems. The review tackles the question of i¾˜How have user-centred evaluations of adaptive and user-adaptive systems been conducted and how can these evaluation practices be improved?' Based on the analysed results of the: (a) evaluation approaches, (b) user-centred evaluation techniques, and (c) evaluation metrics, we propose an evaluation framework for end-user experience in evaluating adaptive systems (EFEx).

Preface

Technology enhanced learning (TEL) aims to design, develop and test socio-technical innovations that will support and enhance learning practices of both individuals and organisations. It is an application domain that generally addresses all types of technology research & development aiming to support teaching and learning activities. Information retrieval is a pivotal activity in TEL, and the deployment of recommender systems has attracted increased interest during the past years. Recommendation methods, techniques and systems open an interesting new approach to facilitate and support learning and teaching. There are plenty of resources available on the Web, both in terms of digital learning content and people resources (e.g. other learners, experts, tutors) that can be used to facilitate teaching and learning tasks. The challenge is to develop, deploy and evaluate systems that provide learners and teachers with meaningful guidance in order to help identify suitable learning resources from a potentially overwhelming variety of choices. The 1st Workshop on Recommender Systems for Technology Enhanced Learning (RecSysTEL) builds upon the tradition of a series of workshops on Social Information Retrieval for Technology Enhanced Learning (SIRTEL), Context-Aware Recommendation for Learning and Towards User Modelling and Adaptive Systems for All (TUMAS-A)a. RecSysTEL was organised jointly by the 4th ACM Conference on Recommender Systems (RecSys 2010) and the 5th European Conference on Technology Enhanced Learning (EC-TEL 2010), on 29–30 September 2010 in Barcelona, Spain. Its main goal was to bring together researchers and practitioners who are working on topics related to the design, development and testing of recommender systems in educational settings as well as present the current status of research in this area and create cross-disciplinary liaisons between the RecSys and ECTEL communities. Overall, its contributions outline the rich potential of TEL as an application area for recommender systems and identify the challenges of developing such systems in a TEL context.