Artículos de conferencia

Permanent URI for this collection

https://repositorio.uc.cl/handle/11534/23156

Browse

Now showing 1 - 13 of 13

A Developer’s Guide to Building and Testing Accessible Mobile Apps
(Association for Computing Machinery (ACM), 2024) Sandoval Alcocer, Juan Pablo; Merino del Campo, Leonel Alejandro; Fernandez Blanco, Alison; Ravelo Mendez, William; Escobar Velásquez, Camilo; Linares Vásquez, Mario
Mobile applications play a relevant role in users’ daily lives by improving and easing daily processes such as commuting or making financial transactions. The aforementioned interactions enhance the usability of commonly used services. Nevertheless, the improvements should also consider special execution environments such as weak network connections or special requirements inherited from the user’s condition. Due to this, the design of mobile applications should be driven by improving the user experience. This tutorial targets the usage of inclusive and accessibility design in the development process of mobile apps. Making sure that applications are accessible to all users, regardless of disabilities, is not just about following the law or fulfilling ethical obligations; it is crucial in creating inclusive and fair digital environments. This tutorial will educate participants on accessibility principles and the available tools. They will gain practical experience with specific Android and iOS platform features, as well as become acquainted with state-of-the-art automated and manual testing tools.
A Family of Discrete Kernels for Presmoothing Test Score Distributions
(Springer, 2024) González Burgos, Jorge Andrés; Wiberg, Marie
In the fields of educational measurement and testing, score distributions are often estimated by the sample relative frequency distribution. As many score distributions are discrete and may have irregularities, it has been common practice to use presmoothing techniques to correct for such irregularities of the score distributions. A common way to conduct presmoothing has been to use log-linear models. In this chapter, we introduce a novel class of discrete kernels that can effectively estimate the probability mass function of scores, providing a presmoothing solution. The chapter includes an empirical illustration demonstrating that the proposed discrete kernel estimates perform as well as or better than the existing methods like log-linear models in presmoothing score distributions. The practical implications of this finding are discussed, highlighting the potential benefits of using discrete kernels in educational measurement contexts. Additionally, the chapter identifies several areas for further research, indicating opportunities for advancing the field’s methodology and practices.
Controlling Trajectories with OneButton and Rhythm
(Association for Computing Machinery, 2024) Bellino, Alessio; Rocchesso, Davide
We demonstrate two-dimensional navigation with velocity control on a single button. Users can vary the speed of the controlled object by rhythm tapping, and can control direction by pressing and tilting, releasing the button once the desired rotation is achieved. Feedback is multisensory. Tactile pulses are being delivered at 30-degree intervals during rotation, simulating the detents of a rotary encoder. Simultaneously, a sonic glissando accompanies rotation, raising or lowering pitch according to the change of direction. Absolute positional feedback is provided visually as well as auditorilly, with an intermittent auditory tone whose pitch conveys vertical position, panned to the left or to the right depending on horizontal position. The rhythmic pace corresponds directly to the on-screen element speed, defined by the tapping interval. Participants will be engaged in a target-following task, thus being able to appreciate the precise speed and direction control of the multisensory navigation experience.
Exact classification of nmr spectra from nmr signals
(Institute of Electrical and Electronics Engineers Inc., 2024) Lehmann, Pedro Izquierdo; Xavier, Aline; Andia Kohnenkampf, Marcelo Edgardo; Sing Long Collao, Carlos Alberto
Nuclear magnetic resonance (NMR) spectroscopy is routinely used to study the properties of matter. Therefore, different materials can be classified according to their NMR spectra. However, the NMR spectra cannot be observed directly, and only the NMR signal, which is a sum of complex exponentials, is directly observable in practice. A popular approach to recover the spectrum is to perform harmonic retrieval, i.e., to reconstruct exactly the spectrum from the NMR signal. However, even when this approach fails, the spectrum might still be classified accurately. In this work, we model the spectrum as an atomic measure to study the performance of classifying the spectrum from the NMR signal, and to determine how it degrades in the presence of additive noise and changes in field intensity. Although we focus on NMR signals, our results are broadly applicable to sum-of-exponential signals. We show numerical results illustrating our claims.
Extracting and Encoding: Leveraging Large Language Models and Medical Knowledge to Enhance Radiological Text Representation
(Association for Computational Linguistics (ACL), 2024) Messina Gallardo, Pablo Alfredo; Vidal, René; Parra Santander, Denis Alejandro; Soto, Álvaro; Araujo, Vladimir
Advancing representation learning in specialized fields like medicine remains challenging due to the scarcity of expert annotations for text and images. To tackle this issue, we present a novel two-stage framework designed to extract high-quality factual statements from free-text radiology reports in order to improve the representations of text encoders and, consequently, their performance on various downstream tasks. In the first stage, we propose a Fact Extractor that leverages large language models (LLMs) to identify factual statements from well-curated domain-specific datasets. In the second stage, we introduce a Fact Encoder (CXRFE) based on a BERT model fine-tuned with objective functions designed to improve its representations using the extracted factual data. Our framework also includes a new embedding-based metric (CXRFEScore) for evaluating chest X-ray text generation systems, leveraging both stages of our approach. Extensive evaluations show that our fact extractor and encoder outperform current state-of-the-art methods in tasks such as sentence ranking, natural language inference, and label extraction from radiology reports. Additionally, our metric proves to be more robust and effective than existing metrics commonly used in the radiology report generation literature. The code of this project is available at https://github.com/PabloMessina/CXR-Fact-Encoder.
HealthIUI: Workshop on Intelligent and Interactive Health User Interfaces
(Association for Computing Machinery, 2025) Brusilovsky, Peter; Parra Santander, Denis Alejandro; Rahdari, B.; Raj, S.; Torkamaan, H.
The HealthIUI workshop explores the integration of intelligent user interfaces in health and care, focusing on AI-driven solutions that enhance user engagement, support clinical decision-making, and improve health information access. The workshop brings together experts from human-computer interaction, AI, and healthcare to address challenges such as transparency, usability, and ethical considerations in AI-assisted health applications. Topics covered include generative AI for patient and caregiver support, AI-powered clinical decision support, adaptive visualization for consumer health information, and explainable AI in nursing care. Through paper presentations and discussions, the workshop fosters interdisciplinary collaboration to advance intelligent health interfaces that balance technical innovation with user-centric design principles.
Normalización terminológica en el ámbito de los bienes patrimoniales chilenos: del corpus a la plataforma web (Short Paper)
(2024) Montero Casado, Ignacia Andrea; Tebé Soriano, Carles; Pissolato de Oliveira, Luciana
En esta ponencia se aborda el proceso de normalización terminológica en un corpus monolingüe en español de Chile del ámbito de la conservación-restauración de bienes patrimoniales realizado en el marco del proyecto FONDEF IDeA I+D “Plataforma para la identificación de las alteraciones a los bienes patrimoniales chilenos”. En específico, se explicarán las etapas metodológicas que constituyen el proceso de normalización: 1) constitución del corpus, 2) extracción de información terminológica del corpus, 3) preparación de los materiales para la sesión de normalización, 4) las sesiones de normalización y 5) elaboración del dossier de salida, previas a la etapa de transición que corresponde al traslado de los resultados de la normalización a la plataforma web. El objetivo de este proceso de normalización es obtener términos relativos a conceptos de alteraciones de los bienes patrimoniales chilenos bien delimitados y consensuados por los especialistas del ámbito para la creación de una base de datos terminológica accesible mediante una plataforma web. La singularidad de la normalización llevada a cabo es que, si bien las decisiones finales recaen en los especialistas del ámbito, el proceso para llegar a esas decisiones está guiado por terminólogos. Esta ponencia se centrará en presentar las particularidades y desafíos que surgieron en las diferentes etapas y las soluciones adoptadas en cada una de ellas, y relacionarlas con las características del ámbito disciplinar, del corpus y con los objetivos y necesidades del proyecto
On the Unexpected Effectiveness of Reinforcement Learning for Sequential Recommendation
(ML Research Press, 2024) Labarca Silva, Álvaro; Parra Santander, Denis; Toro Icarte, Rodrigo Andrés
In recent years, Reinforcement Learning (RL) has shown great promise in session-based recommendation. Sequential models that use RL have reached state-of-the-art performance for the Next-item Prediction (NIP) task. This result is intriguing, as the NIP task only evaluates how well the system can correctly recommend the next item to the user, while the goal of RL is to find a policy that optimizes rewards in the long term - sometimes at the expense of suboptimal short-term performance. Then, how can RL improve the system's performance on short-term metrics? This article investigates this question by exploring proxy learning objectives, which we identify as goals RL models might be following, and thus could explain the performance boost. We found that RL - when used as an auxiliary loss - promotes the learning of embeddings that capture information about the user's previously interacted items. Subsequently, we replaced the RL objective with a straightforward auxiliary loss designed to predict the number of items the user interacted with. This substitution results in performance gains comparable to RL. These findings pave the way to improve performance and understanding of RL methods for recommender systems.
Predicting Effective Flexural Stiffness of Timber Concrete Composite Floors with Different Connection Systems
(2024) Cheraghi-Shirazi, N.; Malek, S.; Guindos Bretones, Pablo; Froese, T.
Timber concrete composite (TCC) floors are increasingly becoming popular in the construction of mass timber buildings due to their enhanced flexural stiffness and improved vibration performance compared to timber floors. Predicting the effective flexural stiffness of TCC floors is vital to calculate the floor deflection or frequency accurately. The most popular method in predicting the effective flexural stiffness of TCC floors is the Gamma method outlined in Eurocode 5–Annex B. Although the Gamma method is generally recommended for predicting the effective flexural stiffness of timber composite floors, its accuracy for TCC floors with various connection systems has not been well established. The current study aims to assess the accuracy of the Gamma method for predicting the effective flexural stiffness of TCC floors by conducting a comprehensive literature review. The measured flexural stiffness values, extracted from the literature, show that the Gamma method can predict the flexural stiffness of TCC floors with an average error of approximately 14% if the shear stiffness of the connection has already been determined experimentally. Since shear tests cannot always be conducted to measure the shear stiffness of a specific connection, engineers resort to analytical equations to estimate it. Some commonly used analytical equations to estimate the shear stiffness of a range of connection systems (e.g., screw, glued, and notched connections) are discussed for this purpose. Experiments under different loading conditions are recommended to determine the accuracy of the Gamma method in predicting the flexural stiffness of TCC floors using the shear stiffness estimated by commonly used analytical equations.
Testing the influence of system effects on the lateral response in t-shaped wood frame shear walls
(2025) Valdivieso Cascante, Diego Nicolás; Almazan Campillay, José Luis; Lopez-García González, Diego; Montaño Castañeda, Jairo Alonso; Liel A.B.; Guindos Bretones, Pablo
This paper examines the impact of transverse shear walls (TSW), out-of-plane bending stiffness of diaphragms (FDIA), and axial (gravity) loading (AXL) on the lateral response of strong wood-frame shear walls (SWs) in multistory light frame timber buildings (LFTBs). Experimental tests assessed the lateral cyclic response of T-shaped SW assemblies with and without diaphragms and gravity load. Tests showed that the TSW effect enhances the lateral stiffness and strength but reduces the deformation capacity. The FDIA and AXL effects further influence the stiffness and strength and compensate in part for the reduction of the deformation capacity due to the TSW effect. Diaphragms also made the T-shaped SW response more symmetrical and improved the evolution of secant stiffness, cumulative dissipated energy, and equivalent viscous damping as the lateral drift increases. Numerical analyses of a theoretical building model with T-shaped SWs showed significant reductions in lateral drift and uplift compared to those of Planar SWs alone, highlighting the importance of considering system effects in the seismic design of LFTBs.
Uncertainty quantification of four phenomenological hysteretic timber models
(2025) Chacón De La Cruz Matías, Fernando Nicolás; Guindos Bretones, Pablo
This article evaluates the uncertainty associated with four phenomenological-based hysteretic timber models from the literature: SAWS/MSTEW, DowelType, modified Richard-Abbott, and ASPID. These models can simulate various timber connections and assemblies, addressing behaviors such as pinching, symmetry and asymmetry, strength and stiffness degradation, and low-cycle fatigue. The models were validated against four experimental benchmark timber tests using an optimized parameter identification process for all cases. The study compared the strength capacity, peak displacement, and energy dissipation. Furthermore, three goodness-of-fit metrics were assessed for the force and energy dissipation histories: Normalized Root Mean Square (NRMS) error, Normalized Mean Absolute (NMA) error, and the coefficient of determination R2. Numerical results indicated that all models, except the SAWS model, achieved good strength capacity and total dissipated energy accuracy, with errors of less than 7%. The models also demonstrated a good fit over time, with NRMS and NMA errors of less than 8.54% for the force history and 4.4% for the dissipated energy history, and R2 values that exceeded 83.11% and 97.9% for force and dissipated energy history, respectively. Therefore, in almost all models, the energy dissipation history fits better than the force one.
Validación de una base de datos terminológicos sobre alteraciones de bienes patrimoniales (Short Paper)
(CEUR-WS, 2024) Tebé, Carles; Pissolato, Luciana; Montero, Ignacia
© 2024 Copyright for this paper by its authors.This paper presents the validation process of a new terminology database to be made available to the public by the end of 2024. The terminology database is the main result of the project "Platform for the identification of alterations to Chilean heritage properties" (ID22I10052) granted in September 2022. This is an initiative funded by the National Agency for Research and Development of Chile (ANID) within the category Fondef IDeA I&D 2022, whose objective is to support the co-financing of applied I&D projects that have a strong scientific component for the development of technologies that can lead to new products, processes or services with the potential to have a positive economic and social impact. Therefore, the process of validation of the database prototype by a group of potential users is a key aspect of this applied result before it is put into service. This paper presents the stages of validation of the terminology database and the criteria related to the evaluation of the design and usability aspects as well as the content of the product.
VICTOR VECTORS @ DIPROMATS 2024: Propaganda Detection with LLM Paraphrasing and Machine Translation
(CEUR-WS, 2024) Fernández, Miguel; Ojeda Aguila, Maximiliano Eduardo; Guevara, Lilly; Varela, Diego; Mendoza Rocha, Marcelo Gabriel; Barrón-Cedeno, Alberto
Identifying propaganda in social media posts is an important task that can help to better understand the strategies applied by policy makers and stake holders when trying to convey their message to the general public. We describe our participation in DIPROMATS 2024 Task 1 on the automated detection and characterization of propaganda techniques and narratives from diplomats of major powers. We show an efficient way to utilize Large Language Models (LLMs) to paraphrase a sample of the training instances, to balance the class distribution in the datasets provided by the shared task. Our submission ranked 1st in Subtask-1a in English (ICM score of 0.2123) and 1st in the bilingual evaluation (ICM score of 0.2048). We also achieved top-3 rankings in Spanish and subtasks 1b and 1c.

Browse

Browsing Artículos de conferencia by browse.metadata.fuente "SCOPUS"

Results Per Page

Sort Options